INDEX

Explanations

specific names or titles

oai_token-act-pair · gpt-3.5-turbo

specific titles, names, or proper nouns related to various media and academic works

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

)</

-0.79

ividual

-0.72

 rush

-0.72

------------------------------------------------

-0.70

 aisle

-0.68

++++++++++++++++

-0.67

âĹ¼

-0.66

vironment

-0.65

)/

-0.65

 carrier

-0.65

POSITIVE LOGITS

 Golf

0.78

Obj

0.67

 Travels

0.66

Keys

0.65

Charlie

0.65

Bad

0.65

OTOS

0.65

metadata

0.64

Forward

0.64

 Byte

0.63

Activations Density 0.392%

specific names or titles

specific titles, names, or proper nouns related to various media and academic works

No Comments

No Known Activations