INDEX

Explanations

actions related to physical movement or interaction

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

ortium

-0.68

][

-0.67

xus

-0.61

inct

-0.61

ocaust

-0.60

yne

-0.59

enny

-0.59

esa

-0.58

ague

-0.58

reshold

-0.58

POSITIVE LOGITS

 accordingly

0.95

 himself

0.73

 thereafter

0.71

 therein

0.68

 alike

0.64

 promptly

0.61

 several

0.59

 consequently

0.59

 herself

0.59

 them

0.58

Activations Density 0.232%

actions related to physical movement or interaction

No Comments

No Known Activations