INDEX

Explanations

comparisons indicating superiority or quantity

oai_token-act-pair · gpt-3.5-turbo

comparative statements indicating greater quantities or frequencies

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

uto

-0.75

EMENT

-0.68

ongo

-0.67

ero

-0.67

ainment

-0.66

ISSION

-0.66

ilic

-0.66

YD

-0.66

eri

-0.65

ursive

-0.64

POSITIVE LOGITS

 usual

0.90

average

0.74

 others

0.71

 average

0.70

 comparable

0.68

adelphia

0.68

 bother

0.67

 ever

0.67

mbuds

0.65

etheless

0.65

Activations Density 0.089%

comparisons indicating superiority or quantity

comparative statements indicating greater quantities or frequencies

No Comments

No Known Activations