INDEX

Explanations

executive positions in organizations

oai_token-act-pair · gpt-3.5-turbo

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

nikov

-0.84

Medium

-0.64

Cooldown

-0.64

 ());

-0.64

jiang

-0.64

worldly

-0.61

States

-0.61

Gaza

-0.60

thodox

-0.60

ento

-0.60

POSITIVE LOGITS

 Jeffrey

1.05

 Larry

0.97

 Patrick

0.97

 Alan

0.97

 Steve

0.96

 Bruce

0.95

 John

0.95

 Tony

0.95

 Randy

0.94

 Richard

0.94

Activations Density 0.152%

executive positions in organizations

No Comments

No Known Activations