INDEX

Explanations

actions related to placing and setting down objects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ho

-0.15

ppo

-0.15

lo

-0.15

ext

-0.15

ADF

-0.14

 ruin

-0.14

 Cave

-0.14

 similar

-0.14

↵

-0.14

aho

-0.14

POSITIVE LOGITS

ÙĪØ¯ÛĮ

0.17

 Insecta

0.16

ossa

0.16

à¸§à¸²à¸ĩ

0.15

.precision

0.15

porno

0.15

_ring

0.15

ãĥ³ãĥĪ

0.14

.jackson

0.14

_principal

0.14

Activations Density 0.196%

actions related to placing and setting down objects

No Comments

No Known Activations