INDEX

Explanations

proper nouns or names

oai_token-act-pair · gpt-3.5-turbo

alphanumeric identifiers and brand names, focusing on their presence in a context

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

 Kung

-0.84

Tec

-0.84

 Sabb

-0.82

Mub

-0.81

 sprint

-0.78

 Masquerade

-0.77

 Stab

-0.76

 Kick

-0.73

jac

-0.70

VEL

-0.70

POSITIVE LOGITS

or

1.45

OR

1.27

oris

1.24

ors

1.21

oros

1.20

orian

1.14

orians

1.13

oro

1.09

orf

1.08

oria

1.08

Activations Density 0.213%

proper nouns or names

alphanumeric identifiers and brand names, focusing on their presence in a context

No Comments

No Known Activations