INDEX

Explanations

books and related terms within the text

oai_token-act-pair · gpt-3.5-turbo

mentions of books and their publication details

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

oling

-0.79

cream

-0.77

orthy

-0.73

someone

-0.64

'>

-0.63

AAAA

-0.62

edia

-0.62

 Helpful

-0.62

ppings

-0.61

requisite

-0.60

POSITIVE LOGITS

 ventures

0.74

stores

0.70

 Artemis

0.69

 autobiography

0.69

 Annotations

0.68

 collaborations

0.66

mates

0.66

 Streets

0.64

 adaptations

0.64

 Surv

0.63

Activations Density 0.161%

books and related terms within the text

mentions of books and their publication details

No Comments

No Known Activations