INDEX

Explanations

user-related terms or strings in the text

oai_token-act-pair · gpt-3.5-turbo

references to user identifiers or mentions of users

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

 Baptist

-0.75

 Lutheran

-0.69

amer

-0.65

forth

-0.63

 Hurricanes

-0.63

hovah

-0.63

 Maid

-0.61

erella

-0.59

SourceFile

-0.59

 Vaugh

-0.58

POSITIVE LOGITS

 interface

1.05

 interfaces

1.01

interface

1.00

 Interface

0.88

base

0.87

atical

0.85

Interface

0.83

Agent

0.82

pace

0.79

hook

0.77

Activations Density 0.026%

user-related terms or strings in the text

references to user identifiers or mentions of users

No Comments

No Known Activations