INDEX

Explanations

phrases related to assembling or putting things together

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

 Ename

-0.99

 hairc

-0.98

 Pamph

-0.92

 Whence

-0.91

 Engraved

-0.85

 Stretcher

-0.85

 oleo

-0.84

 unwarran

-0.83

Eft

-0.81

 ecru

-0.79

POSITIVE LOGITS

 pieces

0.98

 puzzle

0.88

 piece

0.82

 piec

0.82

pieces

0.78

 fragments

0.76

 jigsaw

0.74

assembled

0.71

 rompecabezas

0.70

piece

0.69

Activations Density 0.313%

phrases related to assembling or putting things together

No Comments

No Known Activations