INDEX

Explanations

information related to specific terms or names

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

ktop

-0.71

¥µ

-0.68

 destro

-0.65

PDATE

-0.65

ascus

-0.64

pees

-0.61

 mileage

-0.61

isco

-0.60

patch

-0.60

ptions

-0.59

POSITIVE LOGITS

iasis

1.11

yll

1.03

otle

1.03

uania

1.01

urst

0.99

iop

0.98

ttp

0.94

romy

0.94

ril

0.91

gow

0.90

Activations Density 0.024%

information related to specific terms or names

No Comments

No Known Activations