INDEX

Explanations

phrases and structures indicating measurement or quantification

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

horn

-0.17

anc

-0.16

elia

-0.16

estr

-0.16

again

-0.16

eses

-0.16

 intern

-0.15

orig

-0.15

 again

-0.15

erm

-0.15

POSITIVE LOGITS

oling

0.17

 sesso

0.16

RIPT

0.16

okies

0.15

inski

0.15

amarin

0.14

-horizontal

0.14

isposable

0.13

ityEngine

0.13

quel

0.13

Activations Density 0.026%

phrases and structures indicating measurement or quantification

No Comments

No Known Activations