INDEX

Explanations

comparisons indicating a higher degree or level of something

oai_token-act-pair · gpt-3.5-turbo

comparative phrases indicating a higher degree or increased intensity

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

etimes

-0.71

pex

-0.69

æ©

-0.68

xtap

-0.68

ovies

-0.68

ividual

-0.68

heid

-0.66

utions

-0.66

OCK

-0.65

oes

-0.65

POSITIVE LOGITS

 than

1.21

 efficient

0.97

 accurate

0.96

 complicated

0.96

 likely

0.95

 expensive

0.93

 inclined

0.90

 difficult

0.89

 stringent

0.87

 nuanced

0.87

Activations Density 0.101%

comparisons indicating a higher degree or level of something

comparative phrases indicating a higher degree or increased intensity

No Comments

No Known Activations