INDEX

Explanations

questions and uncertainties expressed in text

oai_token-act-pair · gpt-3.5-turbo

questions and expressions of uncertainty

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

cks

-0.76

ixel

-0.72

eware

-0.70

obin

-0.66

ãĥ¼ãĥ«

-0.66

ipment

-0.65

ulner

-0.65

emies

-0.64

ainers

-0.64

ern

-0.62

POSITIVE LOGITS

 else

0.98

 anybody

0.91

 coincidence

0.83

 anyone

0.79

 anymore

0.77

Else

0.70

 exactly

0.68

 bothered

0.64

 misunder

0.64

else

0.62

Activations Density 0.113%

questions and uncertainties expressed in text

questions and expressions of uncertainty

No Comments

No Known Activations