INDEX

Explanations

prepositions and phrases indicating possession or association

oai_token-act-pair · gpt-3.5-turbo

phrases that express comparisons or descriptors relating to characteristics or qualities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

çļ

-0.77

ESE

-0.77

uers

-0.76

Fs

-0.75

HAEL

-0.75

ãĥĭ

-0.74

ULTS

-0.74

ãĥ¼ãĥĨ

-0.72

ãĤº

-0.72

senal

-0.71

POSITIVE LOGITS

 screwed

0.90

 messed

0.89

 neat

0.89

 like

0.89

 ironic

0.89

 fucked

0.85

 weird

0.83

 goofy

0.83

bum

0.81

 funny

0.80

Activations Density 0.056%

prepositions and phrases indicating possession or association

phrases that express comparisons or descriptors relating to characteristics or qualities

No Comments

No Known Activations