INDEX

Explanations

uncertainty or speculation in statements

oai_token-act-pair · gpt-3.5-turbo

negations and terms indicating uncertainty or doubt

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

incarn

-0.77

 embodiments

-0.69

 Transcript

-0.68

 PRES

-0.65

LIB

-0.63

nect

-0.59

 subsistence

-0.58

 embodiment

-0.57

 obey

-0.56

Eid

-0.56

POSITIVE LOGITS

 surprising

1.01

 unreasonable

0.97

 surprise

0.97

 surprised

0.93

 surpr

0.92

 fret

0.86

essim

0.86

 doubt

0.83

 dissu

0.82

loss

0.79

Activations Density 0.226%

uncertainty or speculation in statements

negations and terms indicating uncertainty or doubt

No Comments

No Known Activations