INDEX

Explanations

references to experimental methods and results

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

ahan

-0.18

experimental

-0.18

Experiment

-0.18

hood

-0.18

_experiment

-0.18

erness

-0.17

heid

-0.16

elper

-0.16

lier

-0.16

aska

-0.16

POSITIVE LOGITS

ally

0.35

ations

0.27

ALLY

0.27

ers

0.23

ists

0.22

ative

0.21

å®¤

0.20

ation

0.20

aly

0.20

atively

0.19

Activations Density 0.017%

references to experimental methods and results

No Comments

No Known Activations