INDEX

Explanations

phrases related to suspicion or being suspected

oai_token-act-pair · gpt-3.5-turbo

instances of suspicion or allegations related to criminal activity

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

jri

-0.70

uits

-0.69

ighth

-0.68

skill

-0.68

psey

-0.66

perty

-0.66

ummer

-0.65

ixt

-0.64

lete

-0.64

braska

-0.64

POSITIVE LOGITS

lessly

0.78

 suspect

0.77

 culprit

0.74

 suspects

0.74

 arson

0.67

 guilty

0.66

 suspicious

0.64

 misuse

0.64

 complicity

0.63

 improper

0.63

Activations Density 0.027%

phrases related to suspicion or being suspected

instances of suspicion or allegations related to criminal activity

No Comments

No Known Activations