INDEX

Explanations

references to irony

oai_token-act-pair · gpt-3.5-turbo

instances of the word "irony" and its variations

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

eding

-0.78

league

-0.75

á

-0.75

igree

-0.72

erto

-0.72

artney

-0.71

lished

-0.71

å£«

-0.69

improve

-0.69

Interstitial

-0.67

POSITIVE LOGITS

 irony

1.13

 twist

1.08

 ironic

1.01

 juxtap

0.90

netflix

0.87

Osw

0.86

 paradox

0.82

 mockery

0.80

 wink

0.78

 twists

0.76

Activations Density 0.039%

references to irony

instances of the word "irony" and its variations

No Comments

No Known Activations