INDEX

Explanations

words related to agreements or statements

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "statement" 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 @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 cannibal

-0.67

 turkey

-0.66

sung

-0.65

âĺ

-0.65

 drink

-0.65

 experienced

-0.64

 âĺ

-0.63

Ïī

-0.62

 rotation

-0.60

 âĻ

-0.60

POSITIVE LOGITS

ements

4.98

ement

2.25

EMENT

1.79

ments

1.45

ancies

1.18

ures

1.12

aments

1.09

etimes

1.07

itures

1.06

urers

1.04

Activations Density 0.020%

words related to agreements or statements

occurrences of the word "statement" in various contexts

No Comments

No Known Activations