INDEX

Explanations

geographical locations and conflict-related terms

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

wcsstore

-0.70

Student

-0.70

sych

-0.68

reads

-0.67

 Person

-0.66

Factor

-0.65

Counter

-0.65

Sales

-0.64

colo

-0.64

Office

-0.64

POSITIVE LOGITS

 inhabited

1.18

 valleys

1.14

 southwest

1.12

 bordering

1.11

 villages

1.09

 northwest

1.08

 valley

1.06

 ridge

1.05

lands

1.04

 mountainous

1.04

Activations Density 3.412%

geographical locations and conflict-related terms

No Comments

No Known Activations