INDEX

Explanations

references to political parties, specifically the Democratic and Republican parties

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

naments

-0.15

Ã¡tek

-0.14

çľł

-0.14

bor

-0.14

bred

-0.14

igen

-0.13

imeType

-0.13

gether

-0.13

 normals

-0.13

echa

-0.13

POSITIVE LOGITS

Ð³Ð°Ð»Ñĸ

0.15

HUD

0.15

ilent

0.15

HUD

0.15

ragon

0.15

istrovstvÃŃ

0.14

%%%%

0.14

 Caldwell

0.14

.AI

0.14

oster

0.14

Activations Density 0.017%

references to political parties, specifically the Democratic and Republican parties

No Comments

No Known Activations