INDEX

Explanations

words related to specific given names

oai_token-act-pair · gpt-3.5-turbo

mentions of the names "Penny" and "Kenny."

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

irtual

-0.90

icity

-0.82

atform

-0.80

andowski

-0.78

inen

-0.78

istar

-0.77

itri

-0.75

itech

-0.73

icted

-0.72

enegger

-0.72

POSITIVE LOGITS

 dreadful

0.82

stick

0.82

pole

0.77

 Kong

0.75

wise

0.75

backer

0.74

 Arcade

0.71

Ĺ

0.70

 Dread

0.70

 Omega

0.69

Activations Density 0.057%

words related to specific given names

mentions of the names "Penny" and "Kenny."

No Comments

No Known Activations