INDEX

Explanations

strong language or expletives

oai_token-act-pair · gpt-3.5-turbo

expressions of frustration or emphasis, particularly those containing the words "damn" and "darn."

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

Profile

-0.65

ors

-0.64

ary

-0.61

 arrivals

-0.61

ipl

-0.61

edia

-0.60

 Arts

-0.60

 Types

-0.58

 Include

-0.57

es

-0.57

POSITIVE LOGITS

 damn

3.99

 darn

3.37

 damned

3.22

 goddamn

3.02

 godd

2.27

 Damn

2.26

 fucking

2.05

Damn

1.86

 freaking

1.80

 fuckin

1.71

Activations Density 0.019%

strong language or expletives

expressions of frustration or emphasis, particularly those containing the words "damn" and "darn."

No Comments

No Known Activations