INDEX

Explanations

technical terms related to software dependencies and artifacts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

æĽ²

-0.15

tsky

-0.15

rame

-0.14

anax

-0.14

ocht

-0.14

landa

-0.14

OrFail

-0.14

chten

-0.14

fell

-0.14

 Charge

-0.14

POSITIVE LOGITS

 rolling

0.18

 Rolling

0.18

 append

0.18

append

0.16

 Pattern

0.16

Skeleton

0.16

Pattern

0.15

unc

0.15

alog

0.15

rolling

0.15

Activations Density 0.005%

technical terms related to software dependencies and artifacts

No Comments

No Known Activations