INDEX

Explanations

references to the name "Lex"

oai_token-act-pair · gpt-3.5-turbo

references to a specific name or character, particularly "Lex."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ONEY

-0.72

============

-0.69

quartered

-0.65

OTS

-0.65

 extremes

-0.65

aternity

-0.64

 Rough

-0.62

ailable

-0.61

========

-0.61

Ø©

-0.60

POSITIVE LOGITS

 Luthor

1.31

 Machina

1.13

icon

0.98

mark

0.92

aminer

0.91

ibly

0.90

avier

0.87

ington

0.87

ibility

0.86

andra

0.85

Activations Density 0.056%

references to the name "Lex"

references to a specific name or character, particularly "Lex."

No Comments

No Known Activations