INDEX

Explanations

phrases where the speaker expresses a comparison or likeness between two things

oai_token-act-pair · gpt-3.5-turbo

phrases indicating comparisons or descriptions

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

heid

-0.74

fulness

-0.71

IAL

-0.70

orest

-0.70

jad

-0.67

ajor

-0.66

 Yard

-0.63

ario

-0.62

ower

-0.61

pak

-0.61

POSITIVE LOGITS

 forget

0.80

 luck

0.79

 kinda

0.75

 neat

0.73

 screwed

0.72

 shit

0.70

 messed

0.69

 scary

0.68

 tease

0.67

fun

0.67

Activations Density 0.040%

phrases where the speaker expresses a comparison or likeness between two things

phrases indicating comparisons or descriptions

No Comments

No Known Activations