INDEX

Explanations

verbs or nouns related to writing or documenting

oai_token-act-pair · gpt-3.5-turbo

terms related to writing and storytelling

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

ONSORED

-0.84

iage

-0.79

hesda

-0.78

ãĤ¤ãĥĪ

-0.76

estinal

-0.71

ailability

-0.71

ouple

-0.70

htar

-0.65

roth

-0.64

ovi

-0.64

POSITIVE LOGITS

ophon

0.79

 omission

0.74

 memos

0.71

 narration

0.71

 autobiography

0.68

 messenger

0.64

 deaf

0.64

 memoir

0.64

bern

0.62

pus

0.62

Activations Density 0.736%

verbs or nouns related to writing or documenting

terms related to writing and storytelling

No Comments

No Known Activations