INDEX

Explanations

words related to laws, regulations, or guidelines

oai_token-act-pair · gpt-3.5-turbo

references to rules and regulations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

Bridge

-0.69

Works

-0.68

ollen

-0.66

unk

-0.62

onto

-0.61

awks

-0.61

ivered

-0.60

imb

-0.60

anse

-0.60

 Hopkins

-0.59

POSITIVE LOGITS

 rule

4.07

 Rule

2.90

rule

2.73

Rule

2.54

 rules

2.17

rules

1.90

 ruled

1.83

 Rules

1.80

rul

1.71

Rules

1.60

Activations Density 0.009%

words related to laws, regulations, or guidelines

references to rules and regulations

No Comments

No Known Activations