INDEX

Explanations

phrases related to being grounded or fixed in a specific context

oai_token-act-pair · gpt-3.5-turbo

terms related to being grounded or stability

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

orr

-0.81

uders

-0.75

uscript

-0.69

apter

-0.69

istors

-0.68

inen

-0.67

Jav

-0.65

ocker

-0.65

å¥³

-0.64

itia

-0.63

POSITIVE LOGITS

 grounding

1.02

 grounded

0.83

hog

0.82

rities

0.71

raint

0.71

ility

0.70

rals

0.70

amera

0.69

 Flat

0.68

rame

0.67

Activations Density 0.015%

phrases related to being grounded or fixed in a specific context

terms related to being grounded or stability

No Comments

No Known Activations