INDEX

Explanations

phrases related to decision making and choices

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

brance

-0.88

ptoms

-0.77

ijk

-0.71

oun

-0.71

igi

-0.68

qv

-0.67

peria

-0.66

emic

-0.66

aura

-0.65

issance

-0.64

POSITIVE LOGITS

 wisely

0.88

 chose

0.82

 chooses

0.79

 chosen

0.77

 randomly

0.75

 choose

0.75

 choice

0.73

 choosing

0.73

 choices

0.66

axe

0.66

Activations Density 0.042%

phrases related to decision making and choices

No Comments

No Known Activations