INDEX

Explanations

words related to the term "Str" which could refer to various entities or concepts across multiple contexts

oai_token-act-pair · gpt-3.5-turbo

instances of a specific pattern or token sequence

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

SHIP

-0.69

peat

-0.66

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.65

ciation

-0.63

WARE

-0.61

 Rampage

-0.61

 Regulatory

-0.60

merce

-0.60

 diminishing

-0.60

 judicial

-0.59

POSITIVE LOGITS

atton

1.31

ategy

1.29

ategic

1.20

anded

1.19

icken

1.15

angers

1.11

ained

1.11

atum

1.10

ateg

1.09

onge

1.09

Activations Density 0.030%

words related to the term "Str" which could refer to various entities or concepts across multiple contexts

instances of a specific pattern or token sequence

No Comments

No Known Activations