INDEX

Explanations

phrases related to extremes or superlatives

oai_token-act-pair · gpt-3.5-turbo

phrases that emphasize the superlative or extreme qualities of various subjects

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤ

-0.70

theless

-0.65

itialized

-0.64

pload

-0.63

mbuds

-0.63

 Uriel

-0.62

 Overs

-0.61

ÃĥÃĤ

-0.60

 Recording

-0.59

 Extras

-0.59

POSITIVE LOGITS

 imaginable

0.90

liest

0.89

 bidder

0.88

 possible

0.84

 conceivable

0.81

iary

0.77

 practicable

0.76

 part

0.73

hest

0.72

 portion

0.71

Activations Density 0.167%

phrases related to extremes or superlatives

phrases that emphasize the superlative or extreme qualities of various subjects

No Comments

No Known Activations