INDEX

Explanations

terms related to patents

oai_token-act-pair · gpt-3.5-turbo

words related to entities or items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

side

-0.79

CTV

-0.76

ãĤ¨ãĥ«

-0.75

STON

-0.71

×¨

-0.70

top

-0.69

Anderson

-0.68

 Armstrong

-0.68

protein

-0.67

books

-0.67

POSITIVE LOGITS

ented

0.88

ailed

0.83

ents

0.82

poons

0.80

ertodd

0.79

poon

0.74

ropy

0.73

 rooting

0.73

vul

0.73

itle

0.72

Activations Density 0.013%

terms related to patents

words related to entities or items

No Comments

No Known Activations