INDEX

Explanations

words related to numerical values and measurements

oai_token-act-pair · gpt-3.5-turbo

metrics and measures related to evaluation criteria

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ĪĴ

-0.82

issance

-0.78

ãĥīãĥ©

-0.75

 engulfed

-0.65

owners

-0.63

ãĤ·

-0.63

 shedding

-0.61

APTER

-0.61

 nurture

-0.60

ãĥ©ãĥ³

-0.58

POSITIVE LOGITS

 depending

1.09

 dictated

1.00

 accordingly

1.00

 based

0.97

 arbitrarily

0.94

 discrepancy

0.92

cale

0.91

 fluct

0.90

 vary

0.88

 varies

0.87

Activations Density 0.410%

words related to numerical values and measurements

metrics and measures related to evaluation criteria

No Comments

No Known Activations