INDEX

Explanations

references to a location or event named "Olen" with varying levels of emphasis

oai_token-act-pair · gpt-3.5-turbo

words associated with the concept of 'loss' or 'stolen'

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

senal

-0.66

 Sheet

-0.65

 Proced

-0.63

 tracts

-0.61

 tract

-0.60

heet

-0.59

 Clicker

-0.58

 Minute

-0.58

adish

-0.58

ĩ

-0.57

POSITIVE LOGITS

olen

1.27

vironment

0.93

aughs

0.89

heim

0.88

ews

0.87

izen

0.86

ovo

0.84

quist

0.82

aston

0.82

wald

0.82

Activations Density 0.018%

references to a location or event named "Olen" with varying levels of emphasis

words associated with the concept of 'loss' or 'stolen'

No Known Activations