INDEX

Explanations

references to central or key concepts in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

edReader

-0.16

Ð¾ÑĢÐ¾ÑĤ

-0.15

aroo

-0.15

orning

-0.15

 Meadows

-0.14

ãĥĨãĥ«

-0.14

illard

-0.14

slu

-0.14

ctp

-0.14

ä»°

-0.14

POSITIVE LOGITS

ast

0.17

core

0.16

most

0.16

/core

0.15

ists

0.15

ist

0.15

oba

0.15

addons

0.15

 core

0.14

-core

0.14

Activations Density 0.021%

references to central or key concepts in various contexts

No Comments

No Known Activations