INDEX

Explanations

comparisons indicating superiority or excellence

oai_token-act-pair · gpt-3.5-turbo

comparative phrases indicating superiority or improvement

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

hran

-0.77

urther

-0.74

VT

-0.72

illary

-0.70

 Juda

-0.70

uum

-0.69

ural

-0.67

Import

-0.67

xy

-0.66

Alt

-0.65

POSITIVE LOGITS

average

0.84

abase

0.79

atever

0.77

 ours

0.75

 usual

0.74

 average

0.73

 anything

0.71

 anybody

0.71

atos

0.69

 Sponge

0.69

Activations Density 0.059%

comparisons indicating superiority or excellence

comparative phrases indicating superiority or improvement

No Comments

No Known Activations