INDEX

Explanations

references to crime and accountability issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

éĹ²

-0.15

 boyc

-0.15

oller

-0.15

landa

-0.14

 Doll

-0.14

ipy

-0.14

referrer

-0.14

.debian

-0.14

 Atlantis

-0.14

atz

-0.14

POSITIVE LOGITS

.qual

0.15

osc

0.15

 treatment

0.15

chers

0.14

LD

0.14

reb

0.14

 Greenwood

0.14

 Writes

0.14

fat

0.14

yles

0.13

Activations Density 0.281%

references to crime and accountability issues

No Comments

No Known Activations