INDEX

Explanations

unusual or unexpected punctuation and symbols within text

oai_token-act-pair · gpt-3.5-turbo

question marks and exclamation points indicating rhetorical or dramatic punctuation

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rogen

-0.71

romy

-0.71

rament

-0.70

estation

-0.70

ral

-0.67

oria

-0.64

rab

-0.64

©¶æ

-0.63

zbollah

-0.62

urally

-0.62

POSITIVE LOGITS

ominated

0.74

 srfAttach

0.69

okers

0.67

ittens

0.66

uits

0.66

CLASSIFIED

0.65

catentry

0.64

hops

0.64

aucuses

0.64

ategory

0.63

Activations Density 0.035%

unusual or unexpected punctuation and symbols within text

question marks and exclamation points indicating rhetorical or dramatic punctuation

No Comments

No Known Activations