INDEX

Explanations

phrases containing the word "says"

oai_token-act-pair · gpt-3.5-turbo

statements attributed to various individuals or sources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

TMZ

-0.69

tumblr

-0.67

 cele

-0.66

 Baton

-0.66

 Presidency

-0.64

 stret

-0.60

 patri

-0.60

 twitter

-0.59

 ransom

-0.59

Holy

-0.59

POSITIVE LOGITS

gaard

0.82

heny

0.79

mson

0.78

xus

0.72

sson

0.69

stals

0.69

 Xiang

0.67

odes

0.67

yk

0.66

 Solution

0.66

Activations Density 0.467%

phrases containing the word "says"

statements attributed to various individuals or sources

No Comments

No Known Activations