INDEX

Explanations

detailed descriptions of visual observations

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

instances of the word "look" and its variations in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

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

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_post

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

accompan

-0.81

âĹ¼

-0.73

icipated

-0.72

Failure

-0.62

theless

-0.62

Els

-0.61

iHUD

-0.60

rots

-0.60

 kcal

-0.59

icks

-0.59

POSITIVE LOGITS

 closely

1.13

 carefully

0.99

at

0.99

 elsewhere

0.98

 around

0.97

 deeper

0.97

 closer

0.94

 farther

0.85

 beneath

0.84

 into

0.84

Activations Density 0.058%

detailed descriptions of visual observations

instances of the word "look" and its variations in different contexts

No Comments

No Known Activations