INDEX

Explanations

specific terms related to significant actions or events

oai_token-act-pair · gpt-3.5-turbo

terms related to actions or changes in policy and decisions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

Native

-0.80

ochet

-0.67

Zip

-0.64

zin

-0.64

avis

-0.63

 vomit

-0.62

thood

-0.59

âĹ¼

-0.59

ose

-0.59

zzy

-0.57

POSITIVE LOGITS

 coincides

0.93

 coincided

0.86

 underscores

0.83

 prompted

0.82

 stems

0.79

 stemmed

0.78

 relates

0.76

 comes

0.74

 amounted

0.73

 infuri

0.73

Activations Density 0.313%

specific terms related to significant actions or events

terms related to actions or changes in policy and decisions

No Comments

No Known Activations