INDEX

Explanations

phrases related to trading or exchange

oai_token-act-pair · gpt-3.5-turbo

references to spin-offs 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 @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 tremend

-0.84

oun

-0.71

¶ħ

-0.63

KEY

-0.61

Ĥª

-0.60

¿½

-0.60

 teasp

-0.60

ĻĤ

-0.59

 organize

-0.59

Ĥİ

-0.58

POSITIVE LOGITS

ments

0.86

ishly

0.85

ment

0.85

ices

0.82

ered

0.81

shore

0.79

ense

0.76

off

0.76

MENTS

0.74

oons

0.73

Activations Density 0.013%

phrases related to trading or exchange

references to spin-offs in various contexts

No Comments

No Known Activations