INDEX

Explanations

occurrences of a specific character or name in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

eyse

-0.19

est

-0.16

ease

-0.16

eh

-0.16

anja

-0.16

Ã´ng

-0.15

tsy

-0.15

an

-0.15

typings

-0.14

atz

-0.14

POSITIVE LOGITS

ermo

0.24

eros

0.21

acker

0.20

rock

0.20

Th

0.19

eron

0.19

eres

0.18

oms

0.18

wait

0.18

ê·¹

0.18

Activations Density 0.016%

occurrences of a specific character or name in the text

No Comments

No Known Activations