INDEX

Explanations

references to personal possessions or experiences

oai_token-act-pair · gpt-3.5-turbo

possessive references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

itect

-0.87

 Uriel

-0.77

 apiece

-0.69

etz

-0.67

 Izan

-0.65

arians

-0.65

 Edison

-0.65

ciation

-0.64

 guiIcon

-0.62

 respectively

-0.62

POSITIVE LOGITS

own

1.36

anmar

1.24

ocard

1.14

stery

1.11

riad

1.07

opia

1.07

 favorite

1.05

opic

1.03

 favourite

0.99

 wife

0.96

Activations Density 0.119%

references to personal possessions or experiences

possessive references

No Comments

No Known Activations