INDEX

Explanations

mentions of dining experiences or restaurants

oai_token-act-pair · gpt-3.5-turbo

instances of the word "din" and associated terms that refer to noise or tumultuous environments

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

ledged

-0.78

dale

-0.70

ãĤ¬

-0.68

mA

-0.68

Ther

-0.67

dos

-0.66

 DRAG

-0.66

 Wilkinson

-0.65

ãĤº

-0.65

heads

-0.64

POSITIVE LOGITS

ership

1.32

ers

1.05

umbers

0.97

erers

0.92

vironment

0.87

iferation

0.87

arios

0.85

ateurs

0.83

erer

0.81

ery

0.81

Activations Density 0.047%

mentions of dining experiences or restaurants

instances of the word "din" and associated terms that refer to noise or tumultuous environments

No Comments

No Known Activations