INDEX

Explanations

references to the name "Lex" or "Lex Luthor."

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Lex," particularly in reference to "Lex Luthor."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ailable

-0.78

ONEY

-0.75

============

-0.75

cknow

-0.70

aternity

-0.69

Â¢

-0.66

quartered

-0.65

=]

-0.65

orld

-0.65

¶ħ

-0.63

POSITIVE LOGITS

 Luthor

1.20

icon

1.02

 Machina

1.01

avier

0.98

aminer

0.93

ibly

0.90

Lex

0.87

ibility

0.87

ician

0.86

ington

0.86

Activations Density 0.049%

references to the name "Lex" or "Lex Luthor."

occurrences of the word "Lex," particularly in reference to "Lex Luthor."

No Comments

No Known Activations