INDEX

Explanations

words associated with support and involvement in terrorism or terrorist organizations

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

pletion

-0.80

ortality

-0.76

gae

-0.71

pler

-0.70

imentary

-0.69

pex

-0.66

clamation

-0.66

elight

-0.65

aqu

-0.65

pect

-0.64

POSITIVE LOGITS

 terrorists

1.08

 rebels

1.07

 dictators

1.05

 militias

1.00

 insurgents

0.99

 terrorist

0.98

 separatist

0.98

 murderous

0.98

 separatists

0.96

 jihadists

0.95

Activations Density 0.292%

words associated with support and involvement in terrorism or terrorist organizations

No Comments

No Known Activations