INDEX

Explanations

references to historical or notable landmarks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

æĬĺ

-0.16

acro

-0.15

ialis

-0.15

 recycl

-0.15

enburg

-0.14

wright

-0.14

oppable

-0.14

oday

-0.14

vant

-0.14

vern

-0.13

POSITIVE LOGITS

itten

0.15

agini

0.14

iska

0.14

æĢģ

0.14

æħĭ

0.14

NL

0.14

ementia

0.14

 consequence

0.14

deo

0.14

 gala

0.14

Activations Density 0.002%

references to historical or notable landmarks

No Comments

No Known Activations