INDEX

Explanations

technical terms or alternative names for concepts or entities

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate something is also referred to by another name or known by alternative terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ftime

-0.83

avorite

-0.65

iscal

-0.64

isman

-0.64

ettel

-0.62

isine

-0.62

achev

-0.62

 upkeep

-0.61

uble

-0.61

Oracle

-0.61

POSITIVE LOGITS

ãĤī

0.93

ãĥĥãĥĪ

0.81

Ð»

0.79

=#

0.78

nep

0.76

Ð¼

0.76

å·

0.74

ãĥł

0.73

ãĥ³

0.70

 Occupations

0.70

Activations Density 0.070%

technical terms or alternative names for concepts or entities

phrases that indicate something is also referred to by another name or known by alternative terms

No Comments

No Known Activations