INDEX

Explanations

patterns of numerical and tabular data

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

yne

-0.17

èįī

-0.15

wo

-0.15

ase

-0.15

ãĥ¼ãĤ¹

-0.15

afen

-0.14

duce

-0.14

æĵį

-0.14

lete

-0.14

ure

-0.14

POSITIVE LOGITS

_DEPRECATED

0.15

 Heller

0.15

/Dk

0.15

bil

0.15

CUS

0.14

 ÐĵÐµÑĢ

0.14

ï¸

0.14

ziel

0.14

elage

0.14

rowable

0.13

Activations Density 0.075%

patterns of numerical and tabular data

No Comments

No Known Activations