INDEX

Explanations

words that indicate a strong interest or enthusiasm

oai_token-act-pair · gpt-3.5-turbo

expressions of eagerness or strong interest

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.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

channelAvailability

-0.81

UID

-0.71

ä»£

-0.70

ICAN

-0.69

inventoryQuantity

-0.66

AGE

-0.65

Ranked

-0.63

 Myst

-0.62

misc

-0.62

 Savior

-0.62

POSITIVE LOGITS

ly

1.10

ness

0.92

eful

0.89

emouth

0.83

irection

0.82

mares

0.80

 keen

0.79

 eyed

0.76

est

0.75

edIn

0.74

Activations Density 0.017%

words that indicate a strong interest or enthusiasm

expressions of eagerness or strong interest

No Comments

No Known Activations