INDEX

Explanations

references to horses and horse-related activities or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

eration

-0.18

lain

-0.17

ewis

-0.17

rive

-0.16

elper

-0.16

rvine

-0.16

Ã©ro

-0.15

erset

-0.15

ests

-0.15

ombs

-0.15

POSITIVE LOGITS

hair

0.27

back

0.26

men

0.23

play

0.20

BACK

0.20

power

0.19

 stable

0.19

hoe

0.19

women

0.18

fly

0.18

Activations Density 0.012%

references to horses and horse-related activities or concepts

No Comments

No Known Activations