INDEX

Explanations

references to academic or authoritative sources, particularly encyclopedias

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

 odom

-0.16

 Paper

-0.16

.lesson

-0.14

itage

-0.14

 Dysfunction

-0.14

èĩ

-0.14

 paper

-0.13

ÎºÎµ

-0.13

 Cosmetic

-0.13

udd

-0.13

POSITIVE LOGITS

 Conc

0.17

enc

0.16

 encyclopedia

0.16

 entries

0.16

 entry

0.16

sa

0.16

Enc

0.16

Dictionary

0.15

ENTRY

0.15

-entry

0.15

Activations Density 0.065%

references to academic or authoritative sources, particularly encyclopedias

No Comments

No Known Activations