INDEX

Explanations

symbols and non-standard characters

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

 Dragonbound

-0.70

 responsive

-0.69

brance

-0.67

 dividend

-0.66

 afore

-0.65

 amplification

-0.65

 Shroud

-0.64

ifiers

-0.64

IFIED

-0.64

 additive

-0.62

POSITIVE LOGITS

Ĭ±

1.61

¶ħ

1.50

¨

1.43

§

1.41

®

1.41

ĭ

1.40

²

1.39

¾

1.38

Į

1.36

¥

1.35

Activations Density 0.002%

symbols and non-standard characters

No Comments

No Known Activations