INDEX

Explanations

names of people and places, likely related to journalism or media

oai_token-act-pair · gpt-3.5-turbo

punctuation marks, particularly commas

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

olas

-0.73

 Beir

-0.71

iter

-0.65

 comprom

-0.63

meyer

-0.61

ensical

-0.61

obbies

-0.61

Foot

-0.61

 millenn

-0.60

byter

-0.60

POSITIVE LOGITS

::::::::

0.90

,,,,

0.89

::::

0.87

,,,,,,,,

0.86

Pa

0.80

ãĥŁ

0.77

taboola

0.76

tel

0.72

then

0.69

ãĤ§

0.69

Activations Density 0.033%

names of people and places, likely related to journalism or media

punctuation marks, particularly commas

No Comments

No Known Activations