INDEX

Explanations

references to a specific organization or company, particularly related to WS (World Series or similar contexts)

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

ÐĴÐŀ

-0.17

uno

-0.15

Ð¾ÑĢÑĤÑĥ

-0.15

erala

-0.14

ERSIST

-0.14

etti

-0.14

å¯§

-0.14

edis

-0.14

Assignable

-0.14

POSITIVE LOGITS

aben

0.16

rschein

0.16

htt

0.15

ampus

0.15

remen

0.15

 Sind

0.15

 Weaver

0.15

_VISIBLE

0.14

aha

0.14

enate

0.14

Activations Density 0.016%

references to a specific organization or company, particularly related to WS (World Series or similar contexts)

No Comments

No Known Activations