INDEX

Explanations

positive attributes or achievements

oai_token-act-pair · gpt-3.5-turbo

the word "outstanding" and its variations, often in the context of excellence or high quality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

©¶æ

-0.91

walk

-0.89

olog

-0.84

jah

-0.82

opt

-0.78

borg

-0.78

anguage

-0.76

uters

-0.76

udo

-0.76

aret

-0.75

POSITIVE LOGITS

NESS

0.97

 outstanding

0.94

 warrants

0.85

ances

0.83

 achievement

0.80

 achievements

0.78

 amounts

0.78

 contributions

0.77

 qualities

0.77

 accomplishments

0.76

Activations Density 0.017%

positive attributes or achievements

the word "outstanding" and its variations, often in the context of excellence or high quality

No Comments

No Known Activations