INDEX

Explanations

terms related to detailed technical instructions

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.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

gee

-0.76

 Tycoon

-0.72

oleon

-0.71

igraph

-0.70

tery

-0.69

idium

-0.68

ivari

-0.68

 foes

-0.65

liest

-0.65

affles

-0.64

POSITIVE LOGITS

 fled

1.01

 clothed

0.93

fledged

0.90

 appreciated

0.83

 transitioned

0.81

 utilized

0.80

 stocked

0.80

 qualified

0.79

 operational

0.79

 functioning

0.79

Activations Density 0.022%

terms related to detailed technical instructions

No Comments

No Known Activations