INDEX

Explanations

proper nouns related to companies or individuals

oai_token-act-pair · gpt-3.5-turbo

words and names associated with specific brands or companies

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 nursery

-0.80

hare

-0.77

 Blocks

-0.75

 Caps

-0.74

 Hamp

-0.73

 hashing

-0.72

 marble

-0.70

 Blazers

-0.70

blocks

-0.69

 Bombs

-0.68

POSITIVE LOGITS

É

2.47

Vo

2.29

Vo

1.72

 Alvarez

1.39

Æ

1.39

 voic

1.34

VO

1.24

vo

1.21

 Nguyen

1.12

ËĪ

1.10

Activations Density 0.043%

proper nouns related to companies or individuals

words and names associated with specific brands or companies

No Comments

No Known Activations