INDEX

Explanations

phrases indicating conclusions or implications

oai_token-act-pair · gpt-3.5-turbo

phrases indicating conclusions or implications

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

thumbnails

-0.74

oked

-0.73

iliate

-0.70

EStreamFrame

-0.69

uded

-0.66

ItemImage

-0.64

ulent

-0.63

 Minotaur

-0.61

eatured

-0.61

bots

-0.58

POSITIVE LOGITS

terday

0.84

 goodbye

0.84

hift

0.80

aucus

0.76

forth

0.68

ãĥĨãĤ£

0.67

forward

0.66

 Ridley

0.65

 Sachs

0.65

ratulations

0.64

Activations Density 0.032%

phrases indicating conclusions or implications

phrases indicating conclusions or implications

No Comments

No Known Activations