INDEX

Explanations

phrases related to the speed or acceleration of various actions

oai_token-act-pair · gpt-3.5-turbo

phrases related to speed and velocity

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

gdala

-0.82

ukong

-0.80

illet

-0.71

sonian

-0.71

loo

-0.69

mberg

-0.69

kok

-0.67

 Aires

-0.67

chens

-0.67

iance

-0.66

POSITIVE LOGITS

 pace

1.12

 paced

1.10

 slower

0.87

 clock

0.84

progress

0.82

 progress

0.81

 pend

0.81

 speeds

0.80

 footsteps

0.80

 speeding

0.79

Activations Density 0.310%

phrases related to the speed or acceleration of various actions

phrases related to speed and velocity

No Comments

No Known Activations