INDEX

Explanations

dates indicated with specific years

oai_token-act-pair · gpt-3.5-turbo

counts and measurements related to time and fiscal data

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

rack

-0.69

ãĤ´ãĥ³

-0.68

¬¼

-0.68

minster

-0.64

Laughs

-0.63

Really

-0.62

Kind

-0.61

 extraord

-0.60

rou

-0.60

ella

-0.60

POSITIVE LOGITS

 however

1.00

 meanwhile

0.89

 moreover

0.83

 there

0.78

 fewer

0.77

 according

0.76

there

0.72

 excluding

0.71

 expect

0.71

 respondents

0.71

Activations Density 0.213%

dates indicated with specific years

counts and measurements related to time and fiscal data

No Comments

No Known Activations