INDEX

Explanations

specific words related to locations or places

oai_token-act-pair · gpt-3.5-turbo

specific names of locations, people, or brands

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

anwhile

-0.57

uyomi

-0.56

ģĸ

-0.55

cffffcc

-0.54

umbnails

-0.53

 inherit

-0.52

ãĤ´ãĥ³

-0.51

*/(

-0.51

etheless

-0.50

 dism

-0.50

POSITIVE LOGITS

 Street

0.68

ona

0.64

eland

0.60

inia

0.59

 Pond

0.59

 Avenue

0.58

isha

0.58

 Island

0.57

oko

0.56

 Lake

0.56

Activations Density 0.779%

specific words related to locations or places

specific names of locations, people, or brands

No Comments

No Known Activations