INDEX

Explanations

references to structural or hierarchical organization in programming or software components

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

 Mage

-0.15

efon

-0.14

sume

-0.14

Ð·Ð°Ð±

-0.14

itin

-0.13

 heterogeneous

-0.13

 Ð¿ÐµÑĢÐµÐ¿

-0.13

itm

-0.13

mounted

-0.13

ehler

-0.12

POSITIVE LOGITS

.util

0.16

framework

0.15

jure

0.15

fila

0.14

lyph

0.14

ernel

0.14

iesel

0.14

ORN

0.14

.common

0.14

-util

0.14

Activations Density 0.089%

references to structural or hierarchical organization in programming or software components

No Comments

No Known Activations