INDEX

Explanations

terms related to procedural steps and decision-making processes

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

loff

-0.15

ÄĻk

-0.15

ulus

-0.15

omite

-0.14

ampo

-0.14

lash

-0.14

umes

-0.14

halt

-0.14

ruc

-0.14

 Humanities

-0.14

POSITIVE LOGITS

 Counsel

0.18

mbH

0.16

 Scheme

0.15

soc

0.15

anc

0.15

hm

0.15

jp

0.15

 Meadow

0.15

 counsel

0.14

 purported

0.14

Activations Density 0.180%

terms related to procedural steps and decision-making processes

No Comments

No Known Activations