INDEX

Explanations

references to significant events and milestones

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

_EXTERN

-0.13

 Hayward

-0.12

 Terrace

-0.12

gerald

-0.12

 ----------------------------------------------------------------------------------------------------------------

-0.12

ess

-0.12

 Levy

-0.12

à¤¼à¥į

-0.12

ä½ľèĢħ

-0.12

 inoc

-0.12

POSITIVE LOGITS

eyin

0.17

eyh

0.16

eyse

0.16

 eskort

0.15

IFn

0.15

CTYPE

0.15

nock

0.15

Î³Î¬

0.14

SCII

0.14

nant

0.14

Activations Density 0.018%

references to significant events and milestones

No Comments

No Known Activations