INDEX

Explanations

phrases related to location or spatial relationships

oai_token-act-pair · gpt-3.5-turbo

phrases indicating spatial or contextual relationships

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

!.

-0.87

+.

-0.82

,...

-0.78

.''.

-0.76

!,

-0.72

*.

-0.66

.........

-0.65

.(

-0.65

$.

-0.64

POSITIVE LOGITS

 these

1.04

 this

0.86

pires

0.74

 such

0.73

these

0.60

stood

0.60

tains

0.59

activity

0.58

 THESE

0.57

efficiency

0.56

Activations Density 0.519%

phrases related to location or spatial relationships

phrases indicating spatial or contextual relationships

No Comments

No Known Activations