INDEX

Explanations

terms related to scaling or growth

oai_token-act-pair · gpt-3.5-turbo

terms related to scaling and growth 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 @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

vous

-0.78

algia

-0.68

naire

-0.67

esson

-0.67

uala

-0.66

riad

-0.63

roit

-0.63

 Leilan

-0.63

 Guilty

-0.63

 Lauder

-0.62

POSITIVE LOGITS

 scale

0.90

 scaled

0.87

 scales

0.82

itized

0.82

Factor

0.81

 heights

0.80

 proportions

0.78

 invari

0.76

itzer

0.71

Scale

0.71

Activations Density 0.012%

terms related to scaling or growth

terms related to scaling and growth in various contexts

No Comments

No Known Activations