INDEX

Explanations

occurrences of the character 'x' in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

 nÃ¡

-0.15

Nm

-0.14

thouse

-0.14

å¯¸

-0.14

ritis

-0.14

rowable

-0.14

aces

-0.14

hammer

-0.14

iera

-0.14

igne

-0.14

POSITIVE LOGITS

TI

0.17

amp

0.16

ur

0.14

reste

0.14

owie

0.13

TI

0.13

itches

0.13

iÃ§

0.13

 coron

0.13

 Pont

0.13

Activations Density 0.007%

occurrences of the character 'x' in different contexts

No Comments

No Known Activations