INDEX

Explanations

phrases related to two opposing groups or parties reaching agreements or conflicts

oai_token-act-pair · gpt-3.5-turbo

references to opposing sides in a discussion or conflict

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ogle

-0.61

 quarry

-0.61

ķ

-0.59

DERR

-0.58

stad

-0.58

utory

-0.58

rupted

-0.57

nce

-0.57

pta

-0.57

 Miko

-0.57

POSITIVE LOGITS

 sexes

1.06

 alike

1.02

 simultaneously

1.00

 sides

1.00

 vying

0.92

 equally

0.90

0.89

 parties

0.87

 halves

0.84

 mutually

0.82

Activations Density 0.197%

phrases related to two opposing groups or parties reaching agreements or conflicts

references to opposing sides in a discussion or conflict

No Comments

No Known Activations