INDEX

Explanations

phrases related to location or direction

oai_token-act-pair · gpt-3.5-turbo

the word "out" in various contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

å£«

-0.63

irm

-0.61

ahime

-0.60

 Brach

-0.59

 Protection

-0.59

itor

-0.58

 Decay

-0.57

 Municip

-0.57

tyr

-0.57

otaur

-0.56

POSITIVE LOGITS

fitted

1.12

 loud

1.06

lasting

1.00

ta

0.92

smart

0.91

landish

0.91

number

0.91

fitting

0.88

doors

0.87

crop

0.86

Activations Density 0.063%

phrases related to location or direction

the word "out" in various contexts

No Comments

No Known Activations