INDEX

Explanations

the name "Ross" with a very strong activation

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Ross."

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

rious

-0.74

 à¨

-0.73

brance

-0.69

urated

-0.67

ACTED

-0.67

lder

-0.65

ulhu

-0.64

 conspicuous

-0.64

undai

-0.64

Ø¹

-0.63

POSITIVE LOGITS

bach

1.02

etti

0.96

inson

0.95

olini

0.87

lyn

0.86

aunders

0.86

andowski

0.84

iter

0.80

endale

0.79

ys

0.78

Activations Density 0.019%

the name "Ross" with a very strong activation

mentions of the name "Ross."

No Comments

No Known Activations