INDEX

Explanations

ambiguous or unclear language

oai_token-act-pair · gpt-3.5-turbo

terms related to ambiguity and 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 @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ICAN

-0.70

din

-0.68

tes

-0.68

Reviewer

-0.66

 Trees

-0.66

 Aden

-0.63

 largeDownload

-0.63

 Congratulations

-0.62

Cross

-0.62

iseum

-0.61

POSITIVE LOGITS

ness

0.86

nesses

0.78

 vaguely

0.78

neb

0.77

 vague

0.77

 outlines

0.76

 recollection

0.75

ly

0.74

 excuses

0.74

 outline

0.73

Activations Density 0.014%

ambiguous or unclear language

terms related to ambiguity and lack of clarity

No Comments

No Known Activations