INDEX

Explanations

words related to alphabets, specifically letters of the alphabet

oai_token-act-pair · gpt-3.5-turbo

references to written correspondence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

rolet

-0.72

SOURCE

-0.69

 WATCH

-0.68

amera

-0.68

Latest

-0.67

Abs

-0.66

minster

-0.66

NOR

-0.66

Govern

-0.65

zinski

-0.64

POSITIVE LOGITS

 letters

1.19

mith

1.09

letter

0.96

ername

0.94

 letter

0.92

worms

0.85

Letter

0.84

 cylinders

0.82

 alphabet

0.80

letters

0.80

Activations Density 0.012%

words related to alphabets, specifically letters of the alphabet

references to written correspondence

No Comments

No Known Activations