INDEX

Explanations

phrases related to connections and relationships between entities

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_fs12288-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

12,288

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

-0.58

stall

-0.58

Ħ¢

-0.57

mac

-0.57

 outnumbered

-0.56

yy

-0.56

onew

-0.56

DIS

-0.55

 SPORTS

-0.55

meyer

-0.54

POSITIVE LOGITS

 dots

0.97

 thereto

0.88

 knot

0.83

 together

0.71

linked

0.70

 knots

0.70

heed

0.65

tones

0.64

 disparate

0.64

itud

0.64

Activations Density 0.164%

phrases related to connections and relationships between entities

No Comments

No Known Activations