INDEX

Explanations

instances of the word "news" and related concepts in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

ÑģÑĤÐ²

-0.15

uide

-0.14

fty

-0.14

usercontent

-0.14

ping

-0.14

ULE

-0.14

APTER

-0.14

bling

-0.14

Å¡

-0.14

Tu

-0.14

POSITIVE LOGITS

letters

0.25

flash

0.23

aram

0.20

ARAM

0.19

line

0.18

FLASH

0.17

feed

0.17

rud

0.17

release

0.17

room

0.17

Activations Density 0.020%

instances of the word "news" and related concepts in various contexts

No Comments

No Known Activations