INDEX

Explanations

the word "assume" followed by some sort of context or action

oai_token-act-pair · gpt-3.5-turbo

instances of the word "assume" and its variations

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

psey

-0.75

gom

-0.73

HCR

-0.71

waters

-0.71

cgi

-0.70

sung

-0.70

encer

-0.69

agra

-0.68

odes

-0.67

turn

-0.67

POSITIVE LOGITS

 assume

0.78

 assumed

0.76

 infall

0.74

 incorrectly

0.71

ially

0.69

 ownership

0.68

LY

0.67

 assumes

0.66

 assumption

0.65

 innocence

0.64

Activations Density 0.036%

the word "assume" followed by some sort of context or action

instances of the word "assume" and its variations

No Comments

No Known Activations