INDEX

Explanations

actions related to work, tasks, or processes being conducted

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

ibrator

-0.15

alm

-0.14

enschaft

-0.14

.rdf

-0.14

ekim

-0.13

heels

-0.13

raid

-0.13

á»ĵng

-0.13

irie

-0.13

POSITIVE LOGITS

lein

0.16

 tasks

0.14

aben

0.14

 Hague

0.14

 Governors

0.14

 duties

0.13

leta

0.13

 tests

0.13

utes

0.13

Activations Density 0.168%

actions related to work, tasks, or processes being conducted

No Comments

No Known Activations