INDEX

Explanations

phrases related to exclusion or removal

oai_token-act-pair · gpt-3.5-turbo

references to "the" 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

uncle

-0.84

utical

-0.78

berus

-0.75

PLA

-0.71

racuse

-0.71

ilib

-0.69

imaru

-0.69

ilee

-0.68

osponsors

-0.67

owicz

-0.67

POSITIVE LOGITS

 equation

0.84

 infancy

0.82

 door

0.80

 bounds

0.78

 gate

0.77

 theater

0.76

 closet

0.76

 drawer

0.75

 nutshell

0.74

 womb

0.74

Activations Density 0.089%

phrases related to exclusion or removal

references to "the" in various contexts

No Comments

No Known Activations