INDEX

Explanations

numerical sequences related to contact information or identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

irie

-0.16

esch

-0.16

elow

-0.15

 Spicer

-0.15

lix

-0.15

idor

-0.14

åĶ

-0.14

ERGE

-0.14

entiful

-0.14

ÅĤaw

-0.14

POSITIVE LOGITS

 Bened

0.16

ngle

0.14

NG

0.14

SF

0.14

 Flip

0.14

 Designer

0.14

ogl

0.13

SF

0.13

 designer

0.13

cub

0.13

Activations Density 0.008%

numerical sequences related to contact information or identifiers

No Comments

No Known Activations