INDEX

Explanations

phrases indicating a contrast or a negative condition

oai_token-act-pair · gpt-3.5-turbo

the word "neither" and its variations in context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

enges

-0.66

tri

-0.65

roxy

-0.65

ÙĴ

-0.65

duc

-0.65

ournals

-0.63

agonal

-0.63

Bang

-0.63

enos

-0.63

è¯

-0.62

POSITIVE LOGITS

theless

0.74

llor

0.74

ndra

0.70

 shalt

0.69

lect

0.68

soever

0.67

 overtly

0.67

wegian

0.66

!--

0.66

 necessarily

0.66

Activations Density 0.015%

phrases indicating a contrast or a negative condition

the word "neither" and its variations in context

No Comments

No Known Activations