INDEX

Explanations

words and phrases describing specific attitudes or behaviors related to willingness or refusal

oai_token-act-pair · gpt-3.5-turbo

expressions related to willingness and unwillingness in decision-making or actions

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

umenthal

-0.63

=-=-=-=-=-=-=-=-

-0.55

oppable

-0.55

 Genie

-0.55

ordon

-0.51

poon

-0.51

ugu

-0.51

oval

-0.51

glomer

-0.51

mson

-0.50

POSITIVE LOGITS

to

1.22

 thereto

0.92

To

0.84

to

0.83

To

0.83

 towards

0.83

 toward

0.82

TO

0.81

lessness

0.78

ta

0.77

Activations Density 0.099%

words and phrases describing specific attitudes or behaviors related to willingness or refusal

expressions related to willingness and unwillingness in decision-making or actions

No Comments

No Known Activations