INDEX

Explanations

text related to actions that involve clicking or downloading

oai_token-act-pair · gpt-3.5-turbo

instances of the word "to."

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Klu

-0.68

Dur

-0.68

andowski

-0.65

ELD

-0.65

offensive

-0.62

 Said

-0.61

calling

-0.60

cffffcc

-0.59

oken

-0.58

announced

-0.58

POSITIVE LOGITS

ggles

1.32

 learn

1.18

 avoid

1.14

 maximize

1.13

 find

1.12

 activate

1.12

 customize

1.11

 ensure

1.10

 obtain

1.09

 explore

1.08

Activations Density 0.165%

text related to actions that involve clicking or downloading

instances of the word "to."

No Comments

No Known Activations