INDEX

Explanations

words related to Grand Prix (GP) racing events

oai_token-act-pair · gpt-3.5-turbo

references to "GP," likely indicating various Grand Prix events or general practitioner contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

sett

-0.69

self

-0.64

itional

-0.64

campaign

-0.63

framework

-0.62

elly

-0.62

————————

-0.62

eous

-0.62

ebted

-0.62

amus

-0.61

POSITIVE LOGITS

GP

0.97

 Canaver

0.85

 Prix

0.82

GP

0.82

OTO

0.75

DX

0.70

ixel

0.68

ross

0.68

allery

0.68

 lapt

0.68

Activations Density 0.004%

words related to Grand Prix (GP) racing events

references to "GP," likely indicating various Grand Prix events or general practitioner contexts

No Comments

No Known Activations