INDEX

Explanations

words related to trust and confidence in various contexts

oai_token-act-pair · gpt-3.5-turbo

expressions related to trust

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

pole

-0.87

 Coulter

-0.68

ois

-0.66

peak

-0.65

İ

-0.61

produced

-0.61

outher

-0.60

RNA

-0.60

Ã¨

-0.58

idding

-0.58

POSITIVE LOGITS

 trust

4.03

trust

2.83

 trusts

2.79

 Trust

2.42

 trusting

2.34

Trust

2.30

 trusted

2.04

 confidence

1.87

 mistrust

1.79

 distrust

1.79

Activations Density 0.018%

words related to trust and confidence in various contexts

expressions related to trust

No Comments

No Known Activations