INDEX

Explanations

instances of the word "what" followed by a strong activation value

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "in what," indicating specific contexts or scenarios

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

orest

-0.65

entry

-0.63

 Memor

-0.59

hig

-0.59

ocket

-0.57

oint

-0.57

achment

-0.56

ovo

-0.56

ridges

-0.56

 guiActiveUnfocused

-0.55

POSITIVE LOGITS

 amounted

1.04

 amounts

1.00

 appears

0.99

 organisers

0.91

 transpired

0.91

 seems

0.89

 seemed

0.86

 resembles

0.84

 became

0.84

 appeared

0.84

Activations Density 0.039%

instances of the word "what" followed by a strong activation value

instances of the phrase "in what," indicating specific contexts or scenarios

No Comments

No Known Activations