INDEX

Explanations

references to administrative or procedural terms related to organizations and documentation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

cheme

-0.16

ivre

-0.15

oze

-0.14

åħ¬åħ±

-0.14

ocytes

-0.14

allback

-0.14

.xtext

-0.14

ACHER

-0.14

ayi

-0.13

chema

-0.13

POSITIVE LOGITS

hiro

0.20

0.17

 Germ

0.17

0.15

0.14

prd

0.14

.face

0.14

 Hogan

0.14

ship

0.14

Activations Density 0.116%

references to administrative or procedural terms related to organizations and documentation

No Comments

No Known Activations