INDEX

Explanations

high-impact actions or achievements

oai_token-act-pair · gpt-3.5-turbo

verbs indicating actions or outcomes

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

ItemTracker

-0.71

wikipedia

-0.68

 Seym

-0.66

assetsadobe

-0.66

ansion

-0.66

ãĤ´ãĥ³

-0.65

pmwiki

-0.64

catentry

-0.64

actionDate

-0.64

Reviewer

-0.63

POSITIVE LOGITS

 Ratio

0.65

 purposes

0.64

ardless

0.61

().

0.60

 them

0.59

 eligible

0.58

ings

0.57

him

0.57

eline

0.57

inton

0.56

Activations Density 0.437%

high-impact actions or achievements

verbs indicating actions or outcomes

No Comments

No Known Activations