INDEX

Explanations

phrases indicating a combination or summary of multiple elements or factors

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

den

-0.88

hole

-0.87

holes

-0.87

writers

-0.87

ble

-0.74

emb

-0.73

geons

-0.73

keeper

-0.72

ysis

-0.71

geon

-0.71

POSITIVE LOGITS

 disparate

1.05

 together

0.93

 seamlessly

0.79

ively

0.78

 neatly

0.77

ogether

0.76

 geographically

0.74

 oppos

0.72

arity

0.72

 bargaining

0.71

Activations Density 2.313%

phrases indicating a combination or summary of multiple elements or factors

No Comments

No Known Activations