INDEX

Explanations

references to behind the scenes action or events

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

 unspeak

-1.12

 impra

-1.01

 withal

-1.01

 tolerably

-1.00

 apprehen

-0.96

 gaily

-0.96

 vainly

-0.94

 impelled

-0.93

 Shakspeare

-0.92

 indescri

-0.91

POSITIVE LOGITS

 felicità

0.75

 pamamagitan

0.74

 panahon

0.73

 bahay

0.72

frastructure

0.71

 behind

0.69

 ideolog

0.69

 bawat

0.68

behind

0.68

 sarili

0.67

Activations Density 0.089%

references to behind the scenes action or events

No Comments

No Known Activations