INDEX

Explanations

references to sticks as objects or symbols

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "stick."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 unfocusedRange

-0.87

APH

-0.76

OTO

-0.71

ONT

-0.69

Uz

-0.69

actionDate

-0.68

ãĥ´

-0.67

inators

-0.66

æĹ

-0.65

âĸ¬

-0.64

POSITIVE LOGITS

holder

0.99

holders

0.95

pole

0.95

lihood

0.91

iness

0.91

handle

0.90

handled

0.87

 blender

0.87

tip

0.85

river

0.82

Activations Density 0.016%

references to sticks as objects or symbols

occurrences of the word "stick."

No Comments

No Known Activations