INDEX

Explanations

connecting words that suggest contrast, condition, or explanation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ses

-0.17

egrity

-0.17

etc

-0.15

hl

-0.14

eur

-0.14

aday

-0.14

utors

-0.14

Ø§

-0.14

yms

-0.14

plorer

-0.14

POSITIVE LOGITS

oret

0.24

bidden

0.21

/-

0.21

!--

0.20

oretical

0.19

/--

0.17

icial

0.15

"./

0.15

isy

0.15

jected

0.14

Activations Density 0.163%

connecting words that suggest contrast, condition, or explanation

No Comments

No Known Activations

connecting words that suggest contrast, condition, or explanation

No Comments

No Known Activations