INDEX

Explanations

terms related to different forms of technical and cultural references

oai_token-act-pair · gpt-3.5-turbo

prepositions, particularly the word "of."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 passers

-0.77

 multiplication

-0.64

 prices

-0.62

FW

-0.62

 partName

-0.60

 ridic

-0.60

 indices

-0.59

 indexes

-0.57

 Brah

-0.56

 wrench

-0.55

POSITIVE LOGITS

sky

1.33

rontal

1.25

eatures

1.14

ortunately

1.11

rame

1.10

lav

1.10

unction

1.08

riend

1.06

ield

1.02

redo

1.02

Activations Density 0.037%

terms related to different forms of technical and cultural references

prepositions, particularly the word "of."

No Comments

No Known Activations