INDEX

Explanations

actions or processes that are done automatically or without manual intervention

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "automatically" and related phrases suggesting automated processes

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

yd

-0.69

past

-0.66

horn

-0.65

 highs

-0.64

 tour

-0.62

 Passion

-0.60

 Manufacturing

-0.60

mare

-0.60

united

-0.59

ife

-0.59

POSITIVE LOGITS

 automatically

3.52

 automatic

1.77

 instantly

1.76

 dynamically

1.69

 magically

1.55

 manually

1.47

 implicitly

1.47

 randomly

1.42

 seamlessly

1.40

 electronically

1.32

Activations Density 0.020%

actions or processes that are done automatically or without manual intervention

occurrences of the word "automatically" and related phrases suggesting automated processes

No Comments

No Known Activations