INDEX

Explanations

words related to expressing agreement

oai_token-act-pair · gpt-3.5-turbo

repetitive phrases or transitional words that introduce new ideas or points

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

UF

-0.72

è¦ļéĨĴ

-0.72

mage

-0.71

Engineers

-0.69

rush

-0.67

alysed

-0.67

interstitial

-0.67

ļéĨĴ

-0.66

SourceFile

-0.66

eton

-0.65

POSITIVE LOGITS

 however

0.98

 although

0.90

yes

0.85

sir

0.83

 despite

0.83

 please

0.79

 according

0.78

 unlike

0.77

 though

0.76

 there

0.76

Activations Density 0.130%

words related to expressing agreement

repetitive phrases or transitional words that introduce new ideas or points

No Comments

No Known Activations