INDEX

Explanations

names containing the letters "Nad"

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Nad" and related references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

@#&

-0.65

 è£ı

-0.64

#$

-0.64

-0.63

 beans

-0.63

bably

-0.62

vernment

-0.62

 Beans

-0.62

 Izan

-0.61

ãĤ§

-0.61

POSITIVE LOGITS

uran

0.98

ler

0.93

algia

0.89

ras

0.87

imir

0.85

imer

0.84

uras

0.84

inished

0.83

alon

0.82

al

0.81

Activations Density 0.041%

names containing the letters "Nad"

mentions of the name "Nad" and related references

No Comments

No Known Activations