INDEX

Explanations

mentions of international conflicts and their associated groups or ideologies

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Canaver

-0.90

ãĤ´ãĥ³

-0.89

DCS

-0.79

 Deadline

-0.79

anticipated

-0.69

éĹĺ

-0.68

ETF

-0.68

arthed

-0.67

è£ħ

-0.66

 Whedon

-0.65

POSITIVE LOGITS

 ..."

1.15

.""

1.10

 âĢ¦"

1.08

 immoral

1.08

 barbaric

1.00

)."

0.98

 corrupt

0.97

 raping

0.96

 Sharia

0.91

 rapes

0.91

Activations Density 1.001%

mentions of international conflicts and their associated groups or ideologies

No Comments

No Known Activations