INDEX

Explanations

repetitive mentions of the word "lot," indicating a focus on the abundance or quantity of something

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

 slightly

-0.17

orns

-0.16

apol

-0.16

elden

-0.15

iors

-0.15

hores

-0.15

ors

-0.15

ject

-0.14

alore

-0.14

axter

-0.14

POSITIVE LOGITS

tery

0.22

tering

0.19

ting

0.18

ãĤĵãģ©

0.17

zheimer

0.16

geme

0.16

terr

0.16

ITE

0.16

 Scri

0.16

TA

0.15

Activations Density 0.035%

repetitive mentions of the word "lot," indicating a focus on the abundance or quantity of something

No Comments

No Known Activations