INDEX

Explanations

descriptions or characteristics being attributed to a specific subject

oai_token-act-pair · gpt-3.5-turbo

phrases that involve descriptions or assessments 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

iaries

-0.66

fman

-0.62

erest

-0.61

 overtake

-0.55

 GOODMAN

-0.55

Osw

-0.54

oa

-0.54

 surpassed

-0.53

ias

-0.53

tackle

-0.53

POSITIVE LOGITS

 differently

0.73

iHUD

0.68

 "$:/

0.67

 Situation

0.67

actly

0.66

SourceFile

0.65

 charact

0.65

 plight

0.63

 behavior

0.63

 succinct

0.62

Activations Density 0.244%

descriptions or characteristics being attributed to a specific subject

phrases that involve descriptions or assessments of various subjects

No Comments

No Known Activations