INDEX

Explanations

determiners ('the', 'a', 'all', '') followed by the word 'that'

oai_token-act-pair · gpt-3.5-turbo

phrases that include the word "that."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Peaks

-0.72

 Procedure

-0.66

izont

-0.65

ãĥ³ãĤ¸

-0.65

 Moons

-0.64

Interstitial

-0.63

 Lords

-0.63

 TAMADRA

-0.63

 Patient

-0.62

 Examiner

-0.61

POSITIVE LOGITS

 transpired

1.09

 mattered

1.07

 separates

0.96

 happened

0.93

 entails

0.88

 happens

0.85

 happ

0.83

 ensued

0.81

 distinguishes

0.81

 surrounds

0.80

Activations Density 0.026%

determiners ('the', 'a', 'all', '') followed by the word 'that'

phrases that include the word "that."

No Comments

No Known Activations