INDEX

Explanations

topics or discussions highlighted or emphasized for communication or debate

oai_token-act-pair · gpt-3.5-turbo

repeated references to "talking" or discussions across different contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

metic

-0.88

xus

-0.85

idon

-0.81

pps

-0.79

xual

-0.79

imaru

-0.78

nery

-0.76

bringer

-0.75

ï¸ı

-0.75

yon

-0.74

POSITIVE LOGITS

 corpses

0.76

 Ducks

0.75

 Squid

0.75

bird

0.75

 glob

0.74

 corpse

0.74

 habits

0.74

 mant

0.73

 Lizard

0.71

 ability

0.71

Activations Density 0.201%

topics or discussions highlighted or emphasized for communication or debate

repeated references to "talking" or discussions across different contexts

No Comments

No Known Activations