INDEX

Explanations

phrases related to events or actions that are happening for the first time

oai_token-act-pair · gpt-3.5-turbo

instances of the word "first" indicating new occurrences or milestones

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

lang

-0.73

gery

-0.73

maps

-0.71

Ïī

-0.70

tics

-0.69

errors

-0.69

morph

-0.67

lov

-0.65

ACTION

-0.65

lah

-0.63

POSITIVE LOGITS

 responders

1.07

 baseman

1.00

 foray

0.95

 glimpse

0.86

 installment

0.85

 ever

0.83

 attempt

0.78

 batch

0.78

 lady

0.78

 incarnation

0.75

Activations Density 0.088%

phrases related to events or actions that are happening for the first time

instances of the word "first" indicating new occurrences or milestones

No Comments

No Known Activations