INDEX

Explanations

the word "short" appearing in various contexts

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "short" 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 @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ILLE

-0.89

Magikarp

-0.76

itational

-0.74

IRO

-0.73

iance

-0.71

ICAN

-0.69

iances

-0.68

MH

-0.68

GAN

-0.68

ONSORED

-0.66

POSITIVE LOGITS

sighted

1.14

ening

1.11

ened

1.07

falls

1.03

ener

1.01

lived

0.99

cuts

0.98

eners

0.98

changed

0.93

 stature

0.92

Activations Density 0.032%

the word "short" appearing in various contexts

occurrences of the word "short" in various contexts

No Comments

No Known Activations