INDEX

Explanations

references to mathematical concepts or names

oai_token-act-pair · gpt-3.5-turbo

occurrences of the substring "el" in words

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

ItemTracker

-0.89

 nomine

-0.79

 Debor

-0.69

displayText

-0.67

 eclipse

-0.65

SourceFile

-0.65

 antioxid

-0.65

 conclud

-0.64

uyomi

-0.64

 fortun

-0.62

POSITIVE LOGITS

izabeth

1.15

baum

1.05

uxe

0.95

ounge

0.94

bach

0.92

ibrary

0.92

ength

0.91

ution

0.87

ayer

0.87

oad

0.87

Activations Density 0.026%

references to mathematical concepts or names

occurrences of the substring "el" in words

No Comments

No Known Activations