INDEX

Explanations

punctuation marks, especially commas

oai_token-act-pair · gpt-3.5-turbo

punctuation and conjunctions in sentences

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

acca

-0.79

PsyNetMessage

-0.77

bard

-0.72

uala

-0.71

arez

-0.71

tnc

-0.71

Latin

-0.68

WT

-0.66

PROV

-0.66

Ã±a

-0.64

POSITIVE LOGITS

 nonetheless

1.37

 nevertheless

1.22

etheless

1.17

 alas

0.91

 beware

0.88

 persisted

0.88

 persists

0.87

 darn

0.84

 retains

0.83

 remains

0.79

Activations Density 0.344%

punctuation marks, especially commas

punctuation and conjunctions in sentences

No Comments

No Known Activations