INDEX

Explanations

names of people or characters in a text

oai_token-act-pair · gpt-3.5-turbo

pronouns and their associated forms in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

Ì¶

-0.65

 consolidation

-0.65

eatures

-0.62

drivers

-0.61

 corros

-0.60

%]

-0.59

critical

-0.57

steamapps

-0.56

 unaff

-0.56

norm

-0.56

POSITIVE LOGITS

neau

0.87

oya

0.80

 Pradesh

0.76

nikov

0.75

chuk

0.73

orf

0.73

wu

0.73

 Brothers

0.73

ofer

0.73

ippi

0.72

Activations Density 0.240%

names of people or characters in a text

pronouns and their associated forms in various contexts

No Comments

No Known Activations