INDEX

Explanations

key terms related to specifications and characteristics of objects or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

iction

-0.17

cmc

-0.16

asure

-0.16

 Hamp

-0.16

Han

-0.14

elan

-0.14

Ð¸Ð½Ñĭ

-0.14

ÎµÎ²

-0.14

ÎµÎ¾

-0.14

Gib

-0.13

POSITIVE LOGITS

otas

0.16

pon

0.15

ISIBLE

0.14

Ð¾ÑĢÑĤ

0.14

ÑģÐ°Ð¼

0.14

ifest

0.14

 Blink

0.14

 Ð¼Ð¸Ð»

0.14

_marshall

0.14

tam

0.13

Activations Density 0.010%

key terms related to specifications and characteristics of objects or entities

No Comments

No Known Activations