INDEX

Explanations

words related to different living situations

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

orsi

-0.69

 Flavoring

-0.68

WARD

-0.67

 involving

-0.64

handled

-0.64

verbal

-0.63

ÏĢ

-0.63

binding

-0.62

agascar

-0.62

Results

-0.61

POSITIVE LOGITS

 confines

1.24

 vicinity

1.14

 midst

1.07

 shadows

1.07

 woods

1.06

 same

1.01

 basement

0.97

 comfort

0.95

 shade

0.95

 attic

0.94

Activations Density 0.166%

words related to different living situations

No Comments

No Known Activations