INDEX

Explanations

names of celebrities

oai_token-act-pair · gpt-3.5-turbo

references to awards or notable achievements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

clair

-0.69

interstitial

-0.68

dden

-0.67

 alignment

-0.66

 lateral

-0.64

LOG

-0.64

 removable

-0.63

 phyl

-0.62

 logged

-0.62

igmatic

-0.61

POSITIVE LOGITS

Rus

2.24

 Wins

2.17

Gos

1.83

Won

1.49

Rus

1.17

Bis

1.06

Kaw

1.05

 Cars

1.05

Wars

1.00

 Krish

0.97

Activations Density 0.014%

names of celebrities

references to awards or notable achievements

No Comments

No Known Activations