INDEX

Explanations

words related to cooking processes and equipment

oai_token-act-pair · gpt-3.5-turbo

instances of dramatic or impactful actions and events

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

ĸļ

-0.87

ccording

-0.68

 Shack

-0.62

oday

-0.59

angan

-0.58

vier

-0.58

estine

-0.57

gain

-0.56

eret

-0.56

 Anthrop

-0.55

POSITIVE LOGITS

.","

0.66

 è£ıè

0.61

"},"

0.59

realDonaldTrump

0.58

"],"

0.57

','

0.57

.''.

0.57

aints

0.53

ss

0.52

ly

0.52

Activations Density 0.610%

words related to cooking processes and equipment

instances of dramatic or impactful actions and events

No Comments

No Known Activations