INDEX

Explanations

instances where a person is going or moving from one place to another

oai_token-act-pair · gpt-3.5-turbo

instances of the word "to"

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 ramifications

-0.76

 levied

-0.75

 impacts

-0.72

 behav

-0.72

 implementations

-0.72

ersed

-0.70

 arising

-0.69

 implications

-0.68

irez

-0.67

 payable

-0.65

POSITIVE LOGITS

pload

1.10

 finish

0.97

ilet

0.96

pless

0.94

 celebrate

0.91

wing

0.90

 fulfill

0.86

wered

0.86

 maximize

0.85

asty

0.85

Activations Density 0.719%

instances where a person is going or moving from one place to another

instances of the word "to"

No Comments

No Known Activations