INDEX

Explanations

names of individuals related to political actions or legal matters

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

uality

-0.83

wcsstore

-0.77

apy

-0.77

ãĤ¦ãĤ¹

-0.73

 ILCS

-0.72

chwitz

-0.65

ministic

-0.65

kaya

-0.65

HTML

-0.65

Role

-0.64

POSITIVE LOGITS

 Senators

0.86

 Senator

0.85

 Ellen

0.83

Sen

0.79

 Congressman

0.78

 Julius

0.77

 Christine

0.77

 former

0.76

 Theodore

0.76

 virtue

0.76

Activations Density 0.132%

names of individuals related to political actions or legal matters

No Comments

No Known Activations