INDEX

Explanations

mentions of specific metrics or numerical values

oai_token-act-pair · gpt-3.5-turbo

mentions of metrics or charts related to performance or resistance in various contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ÂŃ

-1.16

—"

-0.97

ÂŃ

-0.93

—

-0.88

 POLITICO

-0.88

"—

-0.84

Enlarge

-0.79

Åį

-0.77

 âĢķ

-0.76

)—

-0.73

POSITIVE LOGITS

 alot

1.60

 doesnt

1.55

 didnt

1.41

 dont

1.40

tho

1.27

 english

1.22

 thats

1.22

 haha

1.21

lol

1.19

 whats

1.17

Activations Density 2.208%

mentions of specific metrics or numerical values

mentions of metrics or charts related to performance or resistance in various contexts

No Comments

No Known Activations