INDEX

Explanations

phrases related to things being appropriate or suitable in a given context

oai_token-act-pair · gpt-3.5-turbo

the concept of appropriateness or suitability 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 @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Ying

-0.65

cipline

-0.65

 scrolls

-0.61

cipl

-0.61

Lak

-0.61

avez

-0.61

wine

-0.60

gebra

-0.59

agic

-0.59

cheat

-0.59

POSITIVE LOGITS

 fitting

1.06

fitting

1.00

fit

0.93

fitted

0.85

 fitted

0.82

 snug

0.81

Fit

0.75

ably

0.74

 attire

0.74

nell

0.74

Activations Density 0.013%

phrases related to things being appropriate or suitable in a given context

the concept of appropriateness or suitability in various contexts

No Comments

No Known Activations