INDEX

Explanations

phrases related to accountability and responsibility

oai_token-act-pair · gpt-3.5-turbo

instances of a specific character or symbol throughout the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 metic

-0.85

 whistle

-0.80

 radar

-0.79

 shroud

-0.77

 carriage

-0.76

 transit

-0.76

 swarm

-0.75

 groom

-0.75

 badge

-0.75

ul

-0.75

POSITIVE LOGITS

these

1.49

that

1.49

when

1.46

they

1.45

better

1.45

while

1.45

say

1.44

little

1.44

there

1.43

the

1.43

Activations Density 0.088%

phrases related to accountability and responsibility

instances of a specific character or symbol throughout the text

No Comments

No Known Activations