INDEX

Explanations

mentions of drops or droplets

oai_token-act-pair · gpt-3.5-turbo

instances of the word "dro" in various contexts

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

 Canary

-0.91

 Trident

-0.72

 Newsp

-0.71

ãĥĨãĤ£

-0.71

 reconciliation

-0.69

 Freedom

-0.69

 à¨

-0.68

 Blessed

-0.67

ãĥ¼ãĥĨ

-0.67

 specificity

-0.67

POSITIVE LOGITS

plets

1.31

plet

1.29

ppers

1.17

pper

1.13

ves

1.06

ppings

1.05

cks

0.97

opa

0.97

ppy

0.94

vable

0.93

Activations Density 0.011%

mentions of drops or droplets

instances of the word "dro" in various contexts

No Comments

No Known Activations