INDEX

Explanations

uncertainty or lack of clarity in texts

oai_token-act-pair · gpt-3.5-turbo

phrases indicating ambiguity or lack of clarity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

endar

-0.88

INT

-0.81

gall

-0.81

die

-0.77

uilding

-0.76

anders

-0.72

ocard

-0.72

ructose

-0.72

tin

-0.71

onest

-0.71

POSITIVE LOGITS

icably

0.74

ly

0.71

ively

0.71

comings

0.70

 chronological

0.69

 detectable

0.66

 hesitation

0.64

 borders

0.64

 ambiguity

0.63

yet

0.63

Activations Density 0.015%

uncertainty or lack of clarity in texts

phrases indicating ambiguity or lack of clarity

No Comments

No Known Activations