INDEX

Explanations

names of people

oai_token-act-pair · gpt-3.5-turbo

proper nouns and terms related to specific individuals or brands

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

netic

-1.06

net

-0.97

eral

-0.94

uel

-0.93

ific

-0.91

eln

-0.90

ois

-0.88

atted

-0.88

arians

-0.86

antic

-0.85

POSITIVE LOGITS

 confir

0.83

 condem

0.81

kson

0.80

 tradem

0.74

 rightful

0.73

 cryst

0.71

 livest

0.70

 precarious

0.70

 pedest

0.70

 millenn

0.69

Activations Density 0.035%

names of people

proper nouns and terms related to specific individuals or brands

No Comments

No Known Activations