INDEX

Explanations

positive adjectives describing things as pleasant or enjoyable

oai_token-act-pair · gpt-3.5-turbo

instances of the word "nice" used in various contexts

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

arians

-0.82

ogens

-0.81

authorized

-0.78

ichen

-0.74

omics

-0.73

arian

-0.72

eligible

-0.71

inant

-0.71

interrupted

-0.69

rained

-0.68

POSITIVE LOGITS

 gesture

0.87

 nice

0.83

 additions

0.80

bye

0.79

 breeze

0.79

 touches

0.77

 little

0.76

bye

0.76

 enough

0.76

Ã±o

0.75

Activations Density 0.030%

positive adjectives describing things as pleasant or enjoyable

instances of the word "nice" used in various contexts

No Comments

No Known Activations