INDEX

Explanations

phrases related to opposition or resistance

oai_token-act-pair · gpt-3.5-turbo

references to comparisons or contrasts, particularly involving the term "against."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ffe

-0.79

ornia

-0.74

Ut

-0.72

shire

-0.72

CTV

-0.69

redo

-0.66

ocument

-0.66

ayette

-0.66

milo

-0.65

enough

-0.65

POSITIVE LOGITS

 backdrop

1.58

 onslaught

0.82

 odds

0.80

 wall

0.77

 evils

0.77

 invaders

0.77

 enemy

0.77

 presumption

0.76

 intruder

0.75

 temptation

0.74

Activations Density 0.220%

phrases related to opposition or resistance

references to comparisons or contrasts, particularly involving the term "against."

No Comments

No Known Activations