INDEX

Explanations

terms related to accessibility, such as "Access" and "Accessible."

oai_token-act-pair · gpt-3.5-turbo

references to the term "Access" 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 @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Kern

-0.69

annis

-0.68

 ejac

-0.63

Lens

-0.63

GGGG

-0.62

GGGGGGGG

-0.61

birds

-0.61

stal

-0.61

stad

-0.61

asm

-0.58

POSITIVE LOGITS

ibility

1.60

ibly

1.37

ible

1.33

ibl

1.18

ory

1.18

ories

1.18

ibles

1.17

ibilities

1.07

IBLE

1.02

ioned

1.02

Activations Density 0.027%

terms related to accessibility, such as "Access" and "Accessible."

references to the term "Access" in various contexts

No Comments

No Known Activations