INDEX

Explanations

statements related to policies, politics, and organizations

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

CCC

-0.79

 Eleven

-0.77

JD

-0.70

paragraph

-0.66

mire

-0.66

cgi

-0.62

 Daniels

-0.61

Reply

-0.60

Posts

-0.60

Rating

-0.60

POSITIVE LOGITS

're

1.37

selves

1.17

 selves

1.08

 themselves

0.99

'll

0.99

've

0.96

zbollah

0.96

atically

0.92

 respective

0.90

are

0.88

Activations Density 2.048%

statements related to policies, politics, and organizations

No Comments

No Known Activations