INDEX

Explanations

terms related to unlocking content or achievements within a game

oai_token-act-pair · gpt-3.5-turbo

terms related to unlocking or gaining access to features or achievements

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

annis

-0.76

odor

-0.64

 Relative

-0.64

nian

-0.62

nia

-0.62

 Gore

-0.60

 Correct

-0.59

geist

-0.59

gio

-0.59

 Abram

-0.58

POSITIVE LOGITS

 unlocking

1.06

 unlocks

1.01

 unlock

0.96

 unlocked

0.95

 Unlock

0.91

 recess

0.87

locking

0.86

ibility

0.82

locked

0.80

locks

0.80

Activations Density 0.008%

terms related to unlocking content or achievements within a game

terms related to unlocking or gaining access to features or achievements

No Comments

No Known Activations