INDEX

Explanations

mentions of the name "Joe" followed by a numerical activation value

oai_token-act-pair · gpt-3.5-turbo

the repeated mention of the name "Joe."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

NESS

-0.91

hips

-0.78

rawdownloadcloneembedreportprint

-0.77

ties

-0.70

ample

-0.68

ancy

-0.67

ioned

-0.65

imental

-0.65

peed

-0.64

seeing

-0.63

POSITIVE LOGITS

 Biden

1.10

 Arpaio

1.07

Rog

0.91

 Russo

0.88

Pes

0.87

 Scarborough

0.87

 Camel

0.85

 Gibbs

0.84

Rao

0.82

ppo

0.80

Activations Density 0.033%

mentions of the name "Joe" followed by a numerical activation value

the repeated mention of the name "Joe."

No Comments

No Known Activations