INDEX

Explanations

mentions of critically acclaimed films

oai_token-act-pair · gpt-3.5-turbo

mentions of films and movies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

cffff

-0.99

azeera

-0.82

unte

-0.70

urances

-0.68

aples

-0.65

ktop

-0.65

sembly

-0.65

 Canaveral

-0.64

cale

-0.63

 ILCS

-0.62

POSITIVE LOGITS

 itself

1.06

wright

0.98

 premiered

0.98

runner

0.88

 adaptation

0.88

writer

0.86

runners

0.84

 premiere

0.83

 revolves

0.80

writers

0.79

Activations Density 0.121%

mentions of critically acclaimed films

mentions of films and movies

No Comments

No Known Activations