INDEX

Explanations

phrases related to creating various types of situations or environments

oai_token-act-pair · gpt-3.5-turbo

instances of phrases indicating the creation of problematic situations or crises

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ked

-0.77

webkit

-0.65

ebus

-0.65

llan

-0.64

married

-0.63

tom

-0.62

haps

-0.61

aceous

-0.60

king

-0.59

 aloud

-0.59

POSITIVE LOGITS

 havoc

1.09

 opportunities

1.02

 opportunity

0.99

 confusion

0.98

 vacancy

0.96

incent

0.95

 uproar

0.95

 chaos

0.94

 incentives

0.93

 friction

0.93

Activations Density 0.173%

phrases related to creating various types of situations or environments

instances of phrases indicating the creation of problematic situations or crises

No Comments

No Known Activations