INDEX

Explanations

words or prefixes related to the idea of an "ideal" state or standard

oai_token-act-pair · gpt-3.5-turbo

references to ideologies or ideals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

enegger

-0.94

iona

-0.85

ichick

-0.85

ufact

-0.78

wagen

-0.74

cffff

-0.71

lished

-0.69

ions

-0.69

é¾

-0.68

ishable

-0.67

POSITIVE LOGITS

gger

1.03

ll

0.91

lli

0.91

ously

0.82

maid

0.82

lla

0.81

 Dhabi

0.81

llo

0.79

llan

0.74

vice

0.71

Activations Density 0.044%

words or prefixes related to the idea of an "ideal" state or standard

references to ideologies or ideals

No Comments

No Known Activations