INDEX

Explanations

terms related to terminations and dismissals from positions

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

Negative Logits

oky

-0.15

 Stein

-0.14

este

-0.14

ersh

-0.14

 Kitt

-0.14

.delay

-0.13

 legitim

-0.13

Coy

-0.13

åħ¥ãĤĬ

-0.13

POSITIVE LOGITS

 khá»ıi

0.16

eeper

0.14

rosso

0.14

GLE

0.14

enson

0.14

uden

0.14

isify

0.14

/loose

0.14

lyph

0.13

hread

0.13

Activations Density 0.047%

terms related to terminations and dismissals from positions

No Known Activations