INDEX

Explanations

words related to television shows, particularly titles and network names

oai_token-act-pair · gpt-3.5-turbo

various prepositions and conjunctions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 cannabin

-0.61

 Paran

-0.60

 Rats

-0.59

 mete

-0.57

 Scion

-0.55

 Citation

-0.54

 Conc

-0.54

 ISBN

-0.54

 Victims

-0.53

 track

-0.53

POSITIVE LOGITS

orage

0.89

 Pradesh

0.89

ilk

0.83

sembly

0.80

ulum

0.79

umo

0.74

iencies

0.74

oko

0.74

heim

0.73

ibo

0.73

Activations Density 0.118%

words related to television shows, particularly titles and network names

various prepositions and conjunctions

No Comments

No Known Activations