INDEX

Explanations

the word "clear" used in various contexts

oai_token-act-pair · gpt-3.5-turbo

expressions that indicate clarity or certainty about a situation

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

alez

-0.85

umbn

-0.85

izons

-0.76

avorite

-0.75

eatures

-0.75

ritical

-0.72

ussion

-0.71

unte

-0.70

otos

-0.68

@#&

-0.67

POSITIVE LOGITS

 Dragonbound

0.79

 enough

0.76

aneously

0.73

 sailing

0.69

rooms

0.68

URE

0.67

ances

0.67

($)

0.64

MX

0.64

 discrepancies

0.63

Activations Density 0.025%

the word "clear" used in various contexts

expressions that indicate clarity or certainty about a situation

No Comments

No Known Activations