INDEX

Explanations

phrases related to confirmation or verification

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "confirmed."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

selves

-0.69

à¨

-0.66

anza

-0.65

usage

-0.65

 Divide

-0.65

nap

-0.64

apers

-0.64

ngth

-0.63

think

-0.63

cycle

-0.63

POSITIVE LOGITS

 confirmed

3.57

confirmed

2.44

 confirm

2.32

 confirms

2.27

 verified

2.10

 confirmation

2.08

 confirming

2.02

 revealed

1.67

 confir

1.61

 validated

1.60

Activations Density 0.022%

phrases related to confirmation or verification

occurrences of the word "confirmed."

No Comments

No Known Activations