INDEX

Explanations

phrases related to movement or progression along a path or timeline

oai_token-act-pair · gpt-3.5-turbo

phrases referring to paths or trajectories over time

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

orth

-0.85

akeru

-0.82

construct

-0.80

iliary

-0.76

iatric

-0.70

thood

-0.70

itaire

-0.69

riks

-0.67

retty

-0.67

ilot

-0.66

POSITIVE LOGITS

 stairs

0.90

 throats

0.85

 weeds

0.83

 slopes

0.80

 freeway

0.79

 aisle

0.78

 peninsula

0.76

 tubes

0.75

 thirds

0.71

 hallway

0.71

Activations Density 0.124%

phrases related to movement or progression along a path or timeline

phrases referring to paths or trajectories over time

No Comments

No Known Activations