INDEX

Explanations

terms related to responses and reactions, particularly in the context of situations or issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

oya

-0.17

ret

-0.17

-0.16

lette

-0.16

ernet

-0.15

ition

-0.15

ulin

-0.14

ystone

-0.14

inois

-0.14

rf

-0.14

POSITIVE LOGITS

ToSelector

0.24

<|begin_of_text|>

0.19

(Response

0.19

ivate

0.19

.sendRedirect

0.18

/response

0.18

.getWriter

0.16

=response

0.15

.setHeader

0.15

Activations Density 0.044%

terms related to responses and reactions, particularly in the context of situations or issues

No Comments

No Known Activations