INDEX

Explanations

specific terms or keywords related to specific topics

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ment

-0.68

 destro

-0.67

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.64

 cannabin

-0.62

nesota

-0.61

lax

-0.60

 exha

-0.59

 grooming

-0.59

wagen

-0.58

 condem

-0.58

POSITIVE LOGITS

ribute

1.44

ARGET

1.34

olerance

1.33

ractor

1.30

ributes

1.30

ribut

1.25

ension

1.22

ruly

1.21

empt

1.20

ensions

1.18

Activations Density 0.379%

specific terms or keywords related to specific topics

No Comments

No Known Activations