INDEX

Explanations

keywords related to subscribers, callers, contributors, and specific actions or interactions involving them

oai_token-act-pair · gpt-3.5-turbo

terms related to subscribers and callers in a context of service usage

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Protector

-0.63

 Catholicism

-0.63

 Cooking

-0.62

Sau

-0.62

 tourism

-0.62

 Crusader

-0.61

 Rite

-0.61

 Centers

-0.61

 Tour

-0.61

 Sciences

-0.60

POSITIVE LOGITS

hip

1.29

 beware

0.86

hips

0.82

unknown

0.79

who

0.79

IUM

0.77

 laureate

0.76

NAME

0.75

 whom

0.74

who

0.72

Activations Density 0.069%

keywords related to subscribers, callers, contributors, and specific actions or interactions involving them

terms related to subscribers and callers in a context of service usage

No Comments

No Known Activations