INDEX

Explanations

titles or names from various domains such as literature, movies, music, etc

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names of movies, shows, and characters

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

));

-0.68

 reliably

-0.61

 Leilan

-0.59

Zur

-0.56

 glance

-0.56

 Weston

-0.56

 follow

-0.56

 predictable

-0.55

 waivers

-0.55

CPI

-0.55

POSITIVE LOGITS

estine

0.93

foundland

0.87

anmar

0.86

etheus

0.81

cellence

0.80

_.

0.79

agate

0.78

tainment

0.76

odore

0.74

usterity

0.73

Activations Density 0.240%

titles or names from various domains such as literature, movies, music, etc

proper nouns, particularly names of movies, shows, and characters

No Comments

No Known Activations