INDEX

Explanations

phrases related to searching or investigating

oai_token-act-pair · gpt-3.5-turbo

instances of the word "look" and its variations as indicators of searching or examining something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

theless

-0.71

Failure

-0.67

âĹ¼

-0.67

âĢ¢âĢ¢

-0.65

kson

-0.61

icipated

-0.58

aired

-0.58

 proceeding

-0.57

Rah

-0.56

 Prohibition

-0.56

POSITIVE LOGITS

ahead

0.82

 elsewhere

0.80

 closely

0.77

 deeper

0.76

 into

0.76

emouth

0.73

 carefully

0.72

behind

0.72

 inside

0.72

raph

0.71

Activations Density 0.052%

phrases related to searching or investigating

instances of the word "look" and its variations as indicators of searching or examining something

No Comments

No Known Activations