INDEX

Explanations

numbers related to measurements or statistics

oai_token-act-pair · gpt-3.5-turbo

numeric data or statistics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

geries

-0.75

unia

-0.73

ãĥ©ãĥ³

-0.68

 comma

-0.62

ãĥŀ

-0.60

agra

-0.57

 compass

-0.57

 droid

-0.56

 este

-0.56

rator

-0.56

POSITIVE LOGITS

 Conversely

0.93

 Similarly

0.90

],[

0.83

 Moreover

0.82

 Furthermore

0.81

theless

0.81

 âĩ

0.81

 Notably

0.78

 Â¶

0.78

â

0.74

Activations Density 0.095%

numbers related to measurements or statistics

numeric data or statistics

No Comments

No Known Activations