INDEX

Explanations

phrases related to policies or advocacy

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

ĸļ

-1.03

ãĤ¼ãĤ¦ãĤ¹

-0.93

 Sins

-0.76

 Halls

-0.72

 Gorge

-0.69

 Twain

-0.68

 similarities

-0.64

 gorge

-0.64

 sinks

-0.64

owship

-0.64

POSITIVE LOGITS

digy

1.52

verbs

1.24

ctor

1.18

actively

1.18

dding

1.16

pelling

1.16

strate

1.15

ccess

1.10

gression

1.06

dig

1.06

Activations Density 0.012%

phrases related to policies or advocacy

No Comments

No Known Activations