INDEX

Explanations

terms related to influence, choices, and the consequences of actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

ÅĤaw

-0.17

aed

-0.16

avier

-0.15

KL

-0.15

laus

-0.15

ãĥ©ãĤ¹

-0.14

PTY

-0.14

ÙİØª

-0.14

PEAT

-0.14

olland

-0.14

POSITIVE LOGITS

umber

0.16

aller

0.15

å·±

0.14

ê³

0.14

nem

0.14

 Assistant

0.14

 Checker

0.14

odash

0.13

 Linh

0.13

Ð¾Ð±Ñĭ

0.13

Activations Density 0.067%

terms related to influence, choices, and the consequences of actions

No Comments

No Known Activations