INDEX

Explanations

instances of the word "mean" with high activation values

oai_token-act-pair · gpt-3.5-turbo

references to statistical measures, particularly the term "mean."

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

dfx

-0.85

DOM

-0.81

taboola

-0.77

ASED

-0.75

@#&

-0.74

thumbnails

-0.73

Newsletter

-0.72

conservancy

-0.70

anon

-0.70

UNCH

-0.70

POSITIVE LOGITS

 spirited

0.92

ings

0.80

erest

0.76

ingly

0.73

ity

0.71

est

0.69

ework

0.68

ening

0.68

ness

0.67

eway

0.66

Activations Density 0.018%

instances of the word "mean" with high activation values

references to statistical measures, particularly the term "mean."

No Comments

No Known Activations