INDEX

Explanations

numeric patterns

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter 'B' in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Gutenberg

-0.77

 glow

-0.64

 dere

-0.63

 Pharaoh

-0.63

por

-0.63

ãĥĻ

-0.61

 impunity

-0.61

rpm

-0.59

 anonymity

-0.58

 Investigative

-0.58

POSITIVE LOGITS

odies

1.14

rief

1.10

ETA

1.04

anned

1.04

ranch

1.03

rows

1.02

oulder

1.01

urt

1.01

amboo

0.99

achel

0.99

Activations Density 0.046%

numeric patterns

occurrences of the letter 'B' in various contexts

No Comments

No Known Activations