INDEX

Explanations

descriptions of people and everyday activities

oai_token-act-pair · gpt-3.5-turbo

emotional expressions and reactions in conversational contexts

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

FB

-0.78

Sources

-0.74

Fact

-0.66

 disadvant

-0.65

redit

-0.64

 donating

-0.64

 donate

-0.64

 effective

-0.63

 detriment

-0.62

cially

-0.62

POSITIVE LOGITS

 glances

1.10

 sigh

1.10

 smiled

1.06

 nodded

1.06

 nodding

1.04

 stares

1.04

 smile

1.04

 frown

1.03

 grin

1.03

 smiles

1.03

Activations Density 1.273%

descriptions of people and everyday activities

emotional expressions and reactions in conversational contexts

No Comments

No Known Activations