INDEX

Explanations

zoomed-in words or references to zooming in actions

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

nces

-0.79

encia

-0.77

rique

-0.76

ãĥĨãĤ£

-0.72

AFTA

-0.72

rase

-0.72

ãĥĩãĤ£

-0.71

ndra

-0.69

ãĤ·ãĥ£

-0.64

 à¨

-0.62

POSITIVE LOGITS

ooming

1.20

oom

1.07

orph

0.95

 Lens

0.93

itized

0.83

 zoom

0.82

osc

0.81

itsch

0.79

 Zoom

0.78

ature

0.78

Activations Density 0.017%

zoomed-in words or references to zooming in actions

No Comments

No Known Activations