INDEX

Explanations

actions related to using or interacting with something

oai_token-act-pair · gpt-3.5-turbo

references to personal experiences and reactions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ahime

-0.78

 recruitment

-0.70

ENCY

-0.69

immigrant

-0.68

 departures

-0.68

roots

-0.68

 Attacks

-0.66

 treaties

-0.65

 abduction

-0.64

 kidnapping

-0.64

POSITIVE LOGITS

 amazed

1.15

 regretted

1.00

 amused

1.00

 marvel

1.00

 thanked

0.99

 laughed

0.99

 regret

0.99

 applause

0.95

 loved

0.93

 afterward

0.91

Activations Density 0.564%

actions related to using or interacting with something

references to personal experiences and reactions

No Comments

No Known Activations