INDEX

Explanations

information related to interviews and conversations with various individuals

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

 raping

-0.68

 inflic

-0.67

matter

-0.67

 hurting

-0.66

inventoryQuantity

-0.62

ãĥĻ

-0.61

)</

-0.60

itialized

-0.60

onna

-0.59

 endanger

-0.59

POSITIVE LOGITS

 undercover

0.88

 VICE

0.77

edom

0.72

 Goodman

0.70

 extensively

0.69

olicy

0.69

odcast

0.69

 Dominic

0.69

 knowledgeable

0.68

 Greenwald

0.68

Activations Density 0.200%

information related to interviews and conversations with various individuals

No Comments

No Known Activations