INDEX

Explanations

information regarding political events or statements

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

LLOW

-0.67

LOCK

-0.56

acea

-0.56

 encyclopedia

-0.56

Champ

-0.55

WAR

-0.55

boy

-0.55

MC

-0.55

MI

-0.55

Bone

-0.55

POSITIVE LOGITS

uddenly

1.11

 suddenly

0.93

 alas

0.82

 reversed

0.80

 morphed

0.79

 abruptly

0.77

 mysteriously

0.77

instead

0.74

 switched

0.73

anwhile

0.73

Activations Density 0.367%

information regarding political events or statements

No Comments

No Known Activations