INDEX

Explanations

mentions of specific events, organizations, or situations involving notable figures

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Kern

-0.77

astically

-0.72

stal

-0.71

balls

-0.69

GGGG

-0.65

annis

-0.64

pload

-0.63

nen

-0.63

 Fenrir

-0.62

LINE

-0.62

POSITIVE LOGITS

ibility

1.42

ible

1.16

ory

1.11

ibly

1.10

ibles

0.98

ibl

0.95

ories

0.95

IBLE

0.92

ibilities

0.84

edin

0.79

Activations Density 0.018%

mentions of specific events, organizations, or situations involving notable figures

No Comments

No Known Activations