INDEX

Explanations

proper nouns related to individuals or characters named "Bryce" and "Lancaster."

oai_token-act-pair · gpt-3.5-turbo

mentions of specific names, particularly "Bryce Lancaster"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

vest

-0.82

eared

-0.81

uel

-0.81

zig

-0.75

ãĥĺ

-0.72

uers

-0.72

tering

-0.70

tered

-0.69

ãĥ³ãĤ¸

-0.69

eer

-0.68

POSITIVE LOGITS

 Bryce

0.98

 Petty

0.83

 Vaugh

0.79

 Seym

0.78

 Canyon

0.76

 Vernon

0.73

 Moff

0.73

 Gibbs

0.73

anyon

0.70

enery

0.70

Activations Density 0.014%

proper nouns related to individuals or characters named "Bryce" and "Lancaster."

mentions of specific names, particularly "Bryce Lancaster"

No Comments

No Known Activations