INDEX

Explanations

phrases related to reading or continuing with a news story

oai_token-act-pair · gpt-3.5-turbo

phrases that encourage or indicate the continuation of reading

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

»Ĵ

-0.71

pora

-0.67

 Gamble

-0.66

 Profession

-0.64

ignt

-0.63

given

-0.62

Ft

-0.61

esville

-0.60

opard

-0.60

 Clown

-0.59

POSITIVE LOGITS

...]

0.82

 aloud

0.72

 âĨĴ

0.72

 toggle

0.66

 ARTICLE

0.62

 columns

0.61

 BELOW

0.61

 Kafka

0.61

 Below

0.61

ahead

0.61

Activations Density 0.011%

phrases related to reading or continuing with a news story

phrases that encourage or indicate the continuation of reading

No Comments

No Known Activations