INDEX

Explanations

phrases related to instructions or commands

oai_token-act-pair · gpt-3.5-turbo

the end of input segments or document boundaries

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

luster

-0.55

Category

-0.53

 Darling

-0.52

èĢ

-0.48

established

-0.48

 Federation

-0.47

 Moroc

-0.47

Lon

-0.46

herer

-0.45

forced

-0.45

POSITIVE LOGITS

 analyse

0.86

 simplify

0.85

 reproduce

0.84

 activate

0.80

 penetrate

0.79

 disable

0.79

 migrate

0.79

 organise

0.79

 enable

0.78

 modify

0.78

Activations Density 0.331%

phrases related to instructions or commands

the end of input segments or document boundaries

No Comments

No Known Activations