INDEX

Explanations

mentions of specific earthquake magnitudes

oai_token-act-pair · gpt-3.5-turbo

references to the magnitude of earthquakes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

inelli

-0.84

tek

-0.76

bones

-0.75

arters

-0.72

ohydrate

-0.72

starter

-0.72

odge

-0.71

mingham

-0.71

riad

-0.71

ublic

-0.71

POSITIVE LOGITS

 magnitude

1.16

 scale

0.90

 proportions

0.80

 equival

0.76

 sensitivity

0.75

 caliber

0.74

 scales

0.73

 impact

0.73

 significance

0.72

animous

0.72

Activations Density 0.023%

mentions of specific earthquake magnitudes

references to the magnitude of earthquakes

No Comments

No Known Activations