INDEX

Explanations

terms related to network devices and routers

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

-0.89

itate

-0.86

 Petr

-0.76

orie

-0.76

 volunt

-0.75

natureconservancy

-0.75

ittee

-0.72

alse

-0.70

itated

-0.68

aghd

-0.68

POSITIVE LOGITS

 router

1.13

 routers

1.13

 Router

1.05

oute

0.87

 connectivity

0.78

 firmware

0.78

 networking

0.76

Proxy

0.76

 Ethernet

0.76

gear

0.72

Activations Density 0.005%

terms related to network devices and routers

No Comments

No Known Activations