INDEX

Explanations

mentions of specific names related to a news event or topic

oai_token-act-pair · gpt-3.5-turbo

mentions of specific individuals, particularly those related to a notable incident or topic

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

-0.84

ngth

-0.66

 carrot

-0.66

UCT

-0.64

 oats

-0.64

APS

-0.64

REE

-0.64

itting

-0.64

redits

-0.64

CLASS

-0.64

POSITIVE LOGITS

 Mate

1.08

vich

0.82

rette

0.80

Vie

0.79

eer

0.73

lette

0.73

atoon

0.73

querade

0.72

ovich

0.71

lda

0.71

Activations Density 0.019%

mentions of specific names related to a news event or topic

mentions of specific individuals, particularly those related to a notable incident or topic

No Comments

No Known Activations