INDEX

Explanations

the word "once" with high activation values

oai_token-act-pair · gpt-3.5-turbo

instances of the word "once"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

alf

-0.76

LCS

-0.76

MSN

-0.74

DIT

-0.74

IELD

-0.73

RIC

-0.72

NRS

-0.72

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

-0.71

 âĸº

-0.71

ãĥ¼

-0.70

POSITIVE LOGITS

 again

0.82

 bitten

0.76

 Upon

0.75

handedly

0.69

falls

0.68

 Bucc

0.66

 glimps

0.66

soever

0.65

 tasted

0.64

 glance

0.63

Activations Density 0.023%

the word "once" with high activation values

instances of the word "once"

No Comments

No Known Activations