INDEX

Explanations

references to journalists and related terms

oai_token-act-pair · gpt-3.5-turbo

references to journalists or reporting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ministic

-0.71

ensions

-0.71

ranged

-0.68

BuyableInstoreAndOnline

-0.67

emonic

-0.67

 Causes

-0.67

plan

-0.67

inence

-0.67

itionally

-0.66

oku

-0.65

POSITIVE LOGITS

 covering

0.93

 journalists

0.91

 Journalists

0.91

 corps

0.88

Journal

0.84

orters

0.82

 reporters

0.82

 reporter

0.81

 journalist

0.77

cov

0.75

Activations Density 0.059%

references to journalists and related terms

references to journalists or reporting

No Comments

No Known Activations