INDEX

Explanations

phrases related to probability or likelihood

oai_token-act-pair · gpt-3.5-turbo

phrases that express probabilities or chances

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

minus

-0.95

 ILCS

-0.88

Reviewer

-0.80

entric

-0.79

geist

-0.76

hesis

-0.73

idelines

-0.72

raq

-0.71

flows

-0.70

bane

-0.69

POSITIVE LOGITS

 obtaining

0.92

 encountering

0.81

 getting

0.79

 dissu

0.79

 owning

0.79

 doom

0.78

 staying

0.76

 achieving

0.75

 locating

0.75

 acquiring

0.74

Activations Density 0.097%

phrases related to probability or likelihood

phrases that express probabilities or chances

No Comments

No Known Activations