INDEX

Explanations

words related to physical actions or events, particularly involving crime, punishment, or medical conditions

oai_token-act-pair · gpt-3.5-turbo

concepts related to critical events or conditions

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

 heterogeneity

-0.49

 looph

-0.48

reditary

-0.45

 Variant

-0.45

UTC

-0.44

Osw

-0.44

etheless

-0.44

 millenn

-0.43

specific

-0.43

ongyang

-0.41

POSITIVE LOGITS

 fame

0.56

('

0.53

î

0.51

 (Â£

0.49

 whilst

0.47

Eva

0.46

 during

0.45

é¾įå¥ĳå£«

0.45

ynthesis

0.44

,...

0.43

Activations Density 1.546%

words related to physical actions or events, particularly involving crime, punishment, or medical conditions

concepts related to critical events or conditions

No Comments

No Known Activations