INDEX

Explanations

phrases related to evaluating or discussing changes and events

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

Negative Logits

PLIED

-0.72

vey

-0.70

irin

-0.69

emouth

-0.68

ngth

-0.67

VICE

-0.65

etooth

-0.64

hun

-0.61

thouse

-0.61

umbn

-0.60

POSITIVE LOGITS

 radically

1.04

 drastically

1.04

 altering

0.93

 dramatically

0.92

 resh

0.92

changes

0.83

 changes

0.81

alter

0.79

 alters

0.78

Changes

0.77

Activations Density 0.320%

phrases related to evaluating or discussing changes and events

No Known Activations