INDEX

Explanations

phrases indicating similarity or comparison

oai_token-act-pair · gpt-3.5-turbo

phrases that denote comparisons and similarities

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

 squash

-0.77

 bang

-0.69

 Beng

-0.64

 Dota

-0.62

 Bund

-0.61

ELY

-0.59

 demolition

-0.59

 neighbourhood

-0.58

 Rumble

-0.58

 Derby

-0.57

POSITIVE LOGITS

chart

0.86

accompan

0.84

æ©Ł

0.82

ctr

0.78

wise

0.77

quartered

0.76

wcs

0.72

forward

0.70

initions

0.70

nown

0.70

Activations Density 0.026%

phrases indicating similarity or comparison

phrases that denote comparisons and similarities

No Comments

No Known Activations