INDEX

Explanations

instances of the word "more" for further information or details

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

xtap

-0.96

ãĥ¼ãĥĨãĤ£

-0.79

ãĥ£

-0.77

pload

-0.75

ufact

-0.75

§

-0.73

IPS

-0.73

-0.71

ãĥĥãĥĪ

-0.69

ãĥ¼ãĥ«

-0.69

POSITIVE LOGITS

 than

1.17

 info

0.98

 details

0.98

natureconservancy

0.94

 specifics

0.90

 detail

0.86

 importantly

0.78

 about

0.78

 particulars

0.77

 information

0.76

Activations Density 0.044%

instances of the word "more" for further information or details

No Comments

No Known Activations