INDEX

Explanations

the word "told" in varied contexts

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "was told" or variations thereof

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

adesh

-0.76

ivot

-0.71

aho

-0.67

ADD

-0.67

aband

-0.65

avez

-0.63

vous

-0.62

Labor

-0.61

ouf

-0.61

ent

-0.60

POSITIVE LOGITS

tale

0.82

 repeatedly

0.77

terday

0.72

by

0.71

 briefings

0.70

 permission

0.69

 orally

0.69

mith

0.68

 tales

0.68

 beforehand

0.67

Activations Density 0.029%

the word "told" in varied contexts

instances of the phrase "was told" or variations thereof

No Comments

No Known Activations