INDEX

Explanations

numbers followed by a time indication like "m" for minutes

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter 'm' in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Accountability

-0.63

 referen

-0.62

FontSize

-0.60

 fringe

-0.59

 lasting

-0.57

 questioning

-0.57

 broth

-0.57

 compliments

-0.56

chnology

-0.56

 pineapple

-0.55

POSITIVE LOGITS

3.55

mn

1.80

mology

1.78

mi

1.70

ms

1.68

mos

1.65

mic

1.64

mu

1.63

mol

1.60

met

1.57

Activations Density 0.074%

numbers followed by a time indication like "m" for minutes

occurrences of the letter 'm' in the text

No Comments

No Known Activations