INDEX

Explanations

references to desires or preferences for specific actions or outcomes

oai_token-act-pair · gpt-3.5-turbo

expressions of desires or requests

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

Pic

-0.75

 Unknown

-0.65

uras

-0.64

checking

-0.63

cas

-0.61

Old

-0.60

inventoryQuantity

-0.60

ritical

-0.59

ones

-0.59

Cas

-0.58

POSITIVE LOGITS

 suffice

0.77

ELY

0.75

ĸļ

0.72

 someday

0.71

:[

0.70

ģĸ

0.68

ezvous

0.68

otide

0.66

VW

0.65

unic

0.64

Activations Density 0.982%

references to desires or preferences for specific actions or outcomes

expressions of desires or requests

No Comments

No Known Activations