INDEX

Explanations

complex terminology and academic jargon

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ZR

-0.15

eci

-0.15

elsing

-0.15

lad

-0.14

eses

-0.14

esk

-0.14

iesel

-0.14

 Paren

-0.13

andler

-0.13

 virgin

-0.13

POSITIVE LOGITS

enga

0.16

umi

0.15

rush

0.15

0.14

 Wheeler

0.14

ived

0.14

IMENT

0.14

amura

0.14

finished

0.14

Authority

0.13

Activations Density 0.006%

complex terminology and academic jargon

No Comments

No Known Activations