INDEX

Explanations

terms related to energy resources and their classifications

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

Recomm

-0.55

Ire

-0.53

Fla

-0.51

 Marlins

-0.50

laughs

-0.50

ZI

-0.50

NAS

-0.49

 Ashes

-0.49

Brother

-0.49

Honest

-0.48

POSITIVE LOGITS

are

1.06

 were

1.02

 aren

0.96

 weren

0.96

 flowed

0.94

 consisted

0.93

 consist

0.92

 originated

0.89

 include

0.88

 tended

0.88

Activations Density 0.221%

terms related to energy resources and their classifications

No Comments

No Known Activations