INDEX

Explanations

references to specific nationalities or ethnic groups involved in historical or geopolitical contexts

oai_token-act-pair · gpt-4o-mini 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

tains

-0.72

antha

-0.70

ľ

-0.68

boards

-0.67

Topics

-0.64

cells

-0.64

tml

-0.62

Smith

-0.60

ļé

-0.59

states

-0.58

POSITIVE LOGITS

 themselves

1.12

1.04

ervative

0.98

hift

0.94

layer

0.93

ervatives

0.92

']

0.89

 invade

0.87

are

0.84

aurus

0.80

Activations Density 0.098%

references to specific nationalities or ethnic groups involved in historical or geopolitical contexts

No Comments

No Known Activations