INDEX

Explanations

phrases indicating prior knowledge or anticipation

oai_token-act-pair · gpt-3.5-turbo

expressions centered around the concept of knowledge or awareness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

pmwiki

-0.76

phrine

-0.73

*/(

-0.72

uries

-0.71

atre

-0.71

orio

-0.69

otion

-0.68

adish

-0.65

ItemTracker

-0.65

deb

-0.64

POSITIVE LOGITS

 beforehand

1.14

 instinctively

0.90

 nothing

0.81

ledged

0.81

lege

0.79

ledge

0.74

nothing

0.73

bones

0.72

ARDS

0.69

 intimately

0.69

Activations Density 0.064%

phrases indicating prior knowledge or anticipation

expressions centered around the concept of knowledge or awareness

No Comments

No Known Activations