INDEX

Explanations

references to places, events, and activities related to a certain setting or situation

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

 kasarigan

-0.61

 للاسماء

-0.58

CloseOperation

-0.53

 espri

-0.53

 minimalis

-0.53

 parteci

-0.52

 credere

-0.51

Földrajzportál

-0.51

TextHelper

-0.51

 يتيمه

-0.50

POSITIVE LOGITS

 therein

0.88

 thereon

0.81

 unwarran

0.78

 impractica

0.74

 inside

0.72

 thereunder

0.70

 unlaw

0.70

 thereupon

0.67

 excru

0.67

 felicity

0.67

Activations Density 0.545%

references to places, events, and activities related to a certain setting or situation

No Comments

No Known Activations