INDEX

Explanations

specific instances of the word "Sharp"

oai_token-act-pair · gpt-3.5-turbo

mentions of the "Sharp" brand or related products and concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

CVE

-0.84

 Valkyrie

-0.78

 Juliet

-0.71

-0.66

romeda

-0.66

roups

-0.65

ghazi

-0.65

izabeth

-0.65

ESH

-0.64

uthor

-0.63

POSITIVE LOGITS

 Sharp

1.09

 Hots

0.77

Sharp

0.76

ened

0.75

edo

0.74

ness

0.73

ening

0.73

itz

0.72

ener

0.70

 Caption

0.68

Activations Density 0.006%

specific instances of the word "Sharp"

mentions of the "Sharp" brand or related products and concepts

No Comments

No Known Activations