INDEX

Explanations

elements and structures that are related to technical representations or coding syntax

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 19-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L19R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.19.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

á»¯

-0.16

STA

-0.16

rawl

-0.15

ritis

-0.15

 Glob

-0.15

-0.14

 kolo

-0.14

ãĥ«ãĤ¯

-0.14

hab

-0.14

era

-0.14

POSITIVE LOGITS

inet

0.21

oves

0.17

LabelText

0.16

ophe

0.15

.scal

0.15

du

0.15

 Downing

0.15

yles

0.14

olu

0.14

emmel

0.14

Activations Density 0.001%

elements and structures that are related to technical representations or coding syntax

No Comments

No Known Activations