INDEX

Explanations

phrases related to legal consequences such as fines, sentences, and rewards

oai_token-act-pair · gpt-3.5-turbo

terms related to legal penalties, fines, and financial transactions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

nu

-0.67

pmwiki

-0.65

']

-0.65

 scripts

-0.61

xious

-0.61

 gluc

-0.60

apps

-0.60

 suspicions

-0.59

cul

-0.58

 Gods

-0.57

POSITIVE LOGITS

 apiece

1.10

 sandwic

0.78

(~

0.77

 compared

0.71

 consisting

0.71

BIP

0.68

($

0.68

 mark

0.67

 slapped

0.67

 laun

0.67

Activations Density 0.326%

phrases related to legal consequences such as fines, sentences, and rewards

terms related to legal penalties, fines, and financial transactions

No Comments

No Known Activations