INDEX

Explanations

actions and verbs related to processes, tasks, and functionalities

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

 guiActiveUnfocused

-0.67

 wont

-0.62

quickShipAvailable

-0.54

 respectively

-0.50

taboola

-0.49

;;;;;;;;;;;;

-0.49

 thereafter

-0.48

 sadly

-0.47

''.

-0.47

.).

-0.46

POSITIVE LOGITS

 oneself

0.84

ulate

0.83

igate

0.77

uate

0.73

enance

0.69

 yourself

0.67

 your

0.66

ively

0.65

matically

0.64

 ourselves

0.64

Activations Density 0.395%

actions and verbs related to processes, tasks, and functionalities

No Comments

No Known Activations