INDEX

Explanations

phrases related to planning or discussing 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

 Huckabee

-0.65

 Declaration

-0.60

 Pants

-0.60

 Samar

-0.60

CSS

-0.59

 Rusty

-0.58

 Unreal

-0.58

sites

-0.57

 exceptions

-0.56

/.

-0.56

POSITIVE LOGITS

 relates

1.03

 progressed

1.00

 progresses

0.99

 transitioned

0.98

 prepares

0.97

 relate

0.92

 unfolds

0.87

 deems

0.86

 develops

0.85

 contemplated

0.85

Activations Density 0.092%

phrases related to planning or discussing actions

No Comments

No Known Activations