INDEX

Explanations

names with a specific structure, particularly those containing "Johann" or similar variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

elder

-0.17

entin

-0.17

inct

-0.17

Andre

-0.16

Ñİ

-0.16

 Mitar

-0.15

INCT

-0.15

/topics

-0.15

indir

-0.14

 Schiff

-0.14

POSITIVE LOGITS

ansson

0.25

athan

0.23

annes

0.20

anness

0.19

cott

0.16

anes

0.16

ansen

0.16

 Batt

0.16

anning

0.15

 Bapt

0.15

Activations Density 0.010%

names with a specific structure, particularly those containing "Johann" or similar variations

No Comments

No Known Activations