INDEX

Explanations

words related to a product lineup

oai_token-act-pair · gpt-3.5-turbo

references to various types of lineups in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

nai

-0.87

bara

-0.86

ricular

-0.82

ryn

-0.77

birth

-0.75

ode

-0.74

mit

-0.72

rir

-0.72

imb

-0.72

versely

-0.71

POSITIVE LOGITS

 lineup

1.36

XI

0.84

 selections

0.81

 rosters

0.74

 assortment

0.72

 assignments

0.71

 scrimmage

0.71

 LINE

0.70

 choices

0.69

 swaps

0.69

Activations Density 0.006%

words related to a product lineup

references to various types of lineups in different contexts

No Comments

No Known Activations