INDEX

Explanations

instances of the word "Us" followed by varying levels of numbers

oai_token-act-pair · gpt-3.5-turbo

occurrences of "us."

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

 confinement

-0.66

 decoration

-0.63

 concession

-0.62

RAG

-0.62

STE

-0.59

Footnote

-0.59

 concentrating

-0.58

 meth

-0.58

ONY

-0.57

 specialization

-0.57

POSITIVE LOGITS

hers

1.34

ages

1.17

ual

1.17

urers

1.13

ername

1.12

agi

1.08

her

1.03

leep

1.02

earchers

1.00

ern

0.98

Activations Density 0.032%

instances of the word "Us" followed by varying levels of numbers

occurrences of "us."

No Comments

No Known Activations