INDEX

Explanations

references to sharks in various contexts, including attacks, movies, and metaphors

oai_token-act-pair · gpt-3.5-turbo

references to sharks

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

ories

-0.80

ISTER

-0.76

haar

-0.71

ISION

-0.70

 Winchester

-0.69

mble

-0.67

dit

-0.67

onsense

-0.66

ijah

-0.65

 Hemp

-0.63

POSITIVE LOGITS

 fins

0.99

 Sharks

0.92

 sharks

0.88

ulic

0.86

mong

0.85

fish

0.83

 shark

0.82

vati

0.82

 Shark

0.81

affe

0.79

Activations Density 0.013%

references to sharks in various contexts, including attacks, movies, and metaphors

references to sharks

No Comments

No Known Activations