INDEX

Explanations

references to names and terminology related to cultural or academic standards

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

olly

-0.15

inct

-0.15

 Ð½Ð°Ðº

-0.14

_UD

-0.14

³

-0.14

ello

-0.14

stroy

-0.13

CODE

-0.13

 rated

-0.13

cion

-0.13

POSITIVE LOGITS

 usage

0.16

 Treat

0.16

 sometimes

0.15

sometimes

0.15

usage

0.15

Usage

0.15

OMET

0.14

ãĥ«ãĥķ

0.14

UsageId

0.14

 senses

0.14

Activations Density 0.099%

references to names and terminology related to cultural or academic standards

No Comments

No Known Activations