INDEX

Explanations

phrases indicating approval or acceptance

oai_token-act-pair · gpt-3.5-turbo

expressions conveying approval or acceptance

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

marine

-0.86

chet

-0.79

hani

-0.77

riber

-0.68

cano

-0.68

ulhu

-0.68

arcity

-0.66

raped

-0.63

bane

-0.62

quin

-0.62

POSITIVE LOGITS

AY

0.83

lahoma

0.78

 okay

0.78

 alright

0.71

ably

0.68

ol

0.66

BUT

0.66

 enough

0.65

ok

0.64

OK

0.64

Activations Density 0.023%

phrases indicating approval or acceptance

expressions conveying approval or acceptance

No Comments

No Known Activations