INDEX

Explanations

references to honesty, transparency, selectivity, and accuracy in different contexts

oai_token-act-pair · gpt-3.5-turbo

phrases indicating demands for honesty and transparency

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

lehem

-0.78

 Gleaming

-0.69

renheit

-0.68

 Incredible

-0.66

 Alive

-0.65

Ago

-0.65

 weakened

-0.65

cellent

-0.64

vind

-0.64

loo

-0.63

POSITIVE LOGITS

 scheduling

1.02

 sourcing

0.99

 selections

0.97

 handling

0.94

 selecting

0.92

 wording

0.91

 regards

0.90

 terminology

0.90

 allocations

0.89

 assigning

0.88

Activations Density 0.407%

references to honesty, transparency, selectivity, and accuracy in different contexts

phrases indicating demands for honesty and transparency

No Comments

No Known Activations