INDEX

Explanations

irregularly contracted words that include an apostrophe

oai_token-act-pair · gpt-3.5-turbo

words and phrases related to negative emotions or harmful actions

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

 Reef

-0.76

 Leilan

-0.73

 Kris

-0.71

KL

-0.69

 Flavoring

-0.69

KS

-0.65

KP

-0.65

 Ruler

-0.64

 Kirin

-0.64

 Ninth

-0.63

POSITIVE LOGITS

vernment

0.98

usterity

0.87

ploy

0.87

ardless

0.81

selves

0.79

acters

0.79

ï¸ı

0.79

actly

0.78

roc

0.78

ancial

0.77

Activations Density 0.192%

irregularly contracted words that include an apostrophe

words and phrases related to negative emotions or harmful actions

No Comments

No Known Activations