INDEX

Explanations

phrases related to challenges or competitions

oai_token-act-pair · gpt-3.5-turbo

references to challenges or competitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

dyl

-0.76

LIN

-0.71

 stock

-0.70

flu

-0.69

rolog

-0.69

riter

-0.68

rogens

-0.68

ĻĤ

-0.67

rigan

-0.66

gone

-0.66

POSITIVE LOGITS

engers

1.10

enged

1.07

Chall

1.06

 Challenge

1.03

enge

1.00

enges

0.93

chall

0.89

naires

0.87

enger

0.86

plings

0.85

Activations Density 0.010%

phrases related to challenges or competitions

references to challenges or competitions

No Comments

No Known Activations