INDEX

Explanations

words related to communication and interaction

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 lele

-0.64

 Palembang

-0.58

 Rued

-0.54

 palab

-0.54

AccessLevel

-0.52

 kram

-0.52

 Sinal

-0.51

 bandeau

-0.51

uhr

-0.51

 alko

-0.51

POSITIVE LOGITS

 hang

1.23

 hangs

1.20

 hanging

1.15

 hung

1.12

 Hang

1.07

hanging

0.99

hang

0.99

Hanging

0.98

 Hanging

0.97

 hanged

0.96

Activations Density 0.072%

words related to communication and interaction

No Comments

No Known Activations